U.S. patent application number 13/221053 was filed with the patent office on 2011-12-22 for method and system for presenting targeted advertisements.
This patent application is currently assigned to Prime Research Alliance E., Inc.. Invention is credited to Charles A. Eldering, Gregory C. Flickinger.
Application Number | 20110313864 13/221053 |
Document ID | / |
Family ID | 44022294 |
Filed Date | 2011-12-22 |
United States Patent
Application |
20110313864 |
Kind Code |
A1 |
Eldering; Charles A. ; et
al. |
December 22, 2011 |
Method and System for Presenting Targeted Advertisements
Abstract
A method and system for presenting targeted advertisements to a
subscriber includes extracting probabilistic information about
subscriber activities from one or more source and processing the
probabilistic information about subscriber activities to generate a
subscriber characterization vector.
Inventors: |
Eldering; Charles A.;
(Doylestown, PA) ; Flickinger; Gregory C.;
(Furlong, PA) |
Assignee: |
Prime Research Alliance E.,
Inc.
|
Family ID: |
44022294 |
Appl. No.: |
13/221053 |
Filed: |
August 30, 2011 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
13109734 |
May 17, 2011 |
|
|
|
13221053 |
|
|
|
|
09591577 |
Jun 9, 2000 |
7949565 |
|
|
13109734 |
|
|
|
|
09204888 |
Dec 3, 1998 |
7150030 |
|
|
09591577 |
|
|
|
|
60183409 |
Feb 18, 2000 |
|
|
|
60190341 |
Mar 16, 2000 |
|
|
|
60196375 |
Apr 12, 2000 |
|
|
|
Current U.S.
Class: |
705/14.66 |
Current CPC
Class: |
H04N 21/25883 20130101;
H04N 21/44222 20130101; G06Q 30/02 20130101; G06Q 30/0269 20130101;
H04N 21/812 20130101; H04N 21/466 20130101; H04H 20/10 20130101;
G06Q 30/0255 20130101; H04N 21/84 20130101; H04N 7/17318 20130101;
G06Q 30/0251 20130101; H04H 60/46 20130101; H04N 21/25891 20130101;
H04N 21/4662 20130101 |
Class at
Publication: |
705/14.66 |
International
Class: |
G06Q 30/00 20060101
G06Q030/00 |
Claims
1. A method of managing an advertisement system, the method
comprising: (a) accessing a plurality of consumer information
records corresponding to a plurality of consumers, each consumer
information record comprising consumer data in at least two
information categories; (b) accessing an advertisement and a
corresponding ad characterization record, the advertisement
characterization record comprising target characteristics for the
at least two information categories; (c) generating an index score
for each information category in each consumer information record
by correlating the consumer data in the at least two information
categories of the consumer information records with the target
characteristics for the at least two information categories in the
ad characterization record; (d) identifying, based on the index
scores, at least one target consumer to receive the
advertisement.
2. The method of claim 1, further comprising: (e) calculating an
estimated consumer interest value for each of the plurality of
consumers by using a correlation coefficient.
3. The method of claim 2, wherein the correlation coefficient is
the Pearson product-moment coefficient.
4. The method of claim 1, wherein the at least one target consumer
is identified by: (i) summing the index scores in each information
category for each consumer to determine a plurality of summed index
scores; (ii) retrieving a predetermined summed index score
corresponding to the ad characterization record; and (iii)
selecting at least one consumer with a summed index score equal to
or greater than the predetermined summed index score.
5. The method of claim 4, wherein the summed index scores reflect a
probability that the corresponding consumer is interested in the
advertisement.
6. The method of claim 1, wherein the index score for each
information category is determined through the use of one or more
previously developed heuristic rules, wherein the previously
developed heuristic rules relate consumer information to the target
characteristics in at least one of the information categories.
7. The method of claim 6, wherein the previously developed
heuristic rules are probabilistic in nature.
8. A method of managing an advertisement system, the method
comprising: (a) accessing a plurality of consumer information
records corresponding to a plurality of consumers, wherein the
plurality of consumer information records include demographic
information and product preference information corresponding to the
respective consumer; (b) generating a plurality of consumer index
scores based on the consumer information records, wherein each of
the consumer index scores corresponds to the respective consumer;
(c) accessing an advertisement and a corresponding advertisement
characterization profile, wherein the advertisement
characterization profile comprises target characteristics of the
corresponding advertisement; (d) generating an advertisement index
score based on the advertisement characterization profile; and (e)
calculating an estimated consumer interest value for each of the
plurality of consumers by correlating the plurality of consumer
index scores with the advertisement index score.
9. The method of claim 8, further comprising: (f) identifying,
based on the estimated consumer interest value, at least one target
consumer to receive the advertisement.
10. The method of claim 8, wherein the calculating the estimated
consumer interest value uses a correlation coefficient.
11. The method of claim 10, wherein the correlation coefficient is
the Pearson product-moment coefficient.
12. The method of claim 8, wherein the index scores reflect a
probability that the corresponding consumer is interested in a
predefined interest category.
13. The method of claim 12, wherein the index score is generated
based on one or more previously developed heuristic rules, wherein
the previously developed heuristic rules relate consumer
information to a target characteristic of the predefined interest
category.
14. The method of claim 13, wherein the previously developed
heuristic rules are probabilistic in nature.
15. A computer program product, comprising a computer usable medium
having a computer readable program code embodied therein, said
computer readable program code adapted for execution on a computer
to implement a method of managing an advertisement system, said
method comprising: (a) accessing a plurality of consumer
information records corresponding to a plurality of consumers, each
consumer information record comprising consumer data in at least
two information categories; (b) accessing an advertisement and a
corresponding ad characterization record, the advertisement
characterization record comprising target characteristics for the
at least two information categories; (c) generating an index score
for each information category in each consumer information record
by correlating the consumer data in the at least two information
categories of the consumer information records with the target
characteristics for the at least two information categories in the
ad characterization record; (d) identifying, based on the index
scores, at least one target consumer to receive the
advertisement.
16. The computer program product of claim 15, further comprising:
(e) calculating an estimated consumer interest value for each of
the plurality of consumers by using a correlation coefficient.
17. The computer program product of claim 16, wherein the
correlation coefficient is the Pearson product-moment
coefficient.
18. The computer program product of claim 15, wherein the at least
one target consumer is identified by: (i) summing the index scores
in each information category for each consumer to determine a
plurality of summed index scores; (ii) retrieving a predetermined
summed index score corresponding to the ad characterization record;
and (iii) selecting at least one consumer with a summed index score
equal to or greater than the predetermined summed index score.
19. An advertisement matching system in a computing environment,
said system comprising: an electronic storage unit configured to
store a plurality of consumer information records, the records
comprising consumer data in at least two information categories; a
receiving unit configured to receive advertisement data including
an advertisement and a corresponding advertisement characterization
record, the advertisement characterization record comprising target
characteristics for the at least two information categories; and a
processor configured to: (i) interpret the consumer information
records and the advertisement characterization record; (ii)
generate an index score for each information category in each
consumer information record by correlating the consumer data in the
at least two information categories of the consumer information
records with the target characteristics for the at least two
information categories in the advertisement characterization
record; and (iii) identify, based on the index scores, at least one
target consumer to receive the advertisement.
20. The advertisement matching system of claim 19, further
comprising: an advertisement delivery unit for delivering the
advertisement to the at least one target consumer.
Description
CROSS REFERENCE TO RELATED APPLICATIONS
[0001] This application is a continuation of U.S. application Ser.
No. 13/109,734, filed May 17, 2011, and entitled Method and System
for Presenting Targeted Advertisements, which is a continuation of
U.S. application Ser. No. 09/591,577, filed Jun. 9, 2000, now U.S.
Pat. No. 7,949,565, and entitled Privacy Protected Advertising
System, which claims the benefit of U.S. Provisional Application
Nos. 60/183,409, filed Feb. 18, 2000, entitled Ad Matching Service;
60/190,341, filed Mar. 16, 2000, entitled Privacy Protected
Filtering and Profiling System; and 60/196,375, filed Apr. 12,
2000, entitled Ad Matching Service. U.S. application Ser. No.
09/591,577 is a Continuation-in-part of U.S. application Ser. No.
09/204,888, filed Dec. 3, 1998, and entitled Subscriber
Characterization System, now U.S. Pat. No. 7,150,030. The entire
disclosures of all of the above applications are incorporated
herein by reference.
[0002] This application is related to U.S. application Ser. No.
09/205,653, filed Dec. 3, 1998, now U.S. Pat. No. 6,457,010; Ser.
No. 09/205,119, filed Dec. 3, 1998; Ser. No. 09/268,519, filed Mar.
12, 1999, now U.S. Pat. No. 6,298,348; Ser. No. 09/268,526, filed
Mar. 12, 1999, now U.S. Pat. No. 6,216,129; and Ser. No.
09/268,520, filed Mar. 12, 1999, now U.S. Pat. No. 6,324,519. All
of the above applications are incorporated herein by reference in
their entirety, but are not admitted to be prior art.
BACKGROUND OF THE INVENTION
[0003] Advertising forms an important part of broadcast programming
including broadcast video (television), radio and printed media.
The revenues generated from advertisers subsidize and in some cases
pay entirely for programming received by subscribers. For example,
over the air broadcast programming (non-cable television) is
provided entirely free to subscribers and is essentially paid for
by the advertisements placed in the shows that are watched. Even in
cable television systems and satellite-based systems, the revenues
from advertisements subsidize the cost of the programming, and were
it not for advertisements, the monthly subscription rates for cable
television would be many times higher than at present. Radio
similarly offers free programming based on payments for
advertising. The low cost of newspapers and magazines is based on
the subsidization of the cost of reporting, printing and
distribution from the advertising revenues.
[0004] Techniques for inserting pre-recorded spot messages into
broadcast transmission have been known. Generally, broadcast video
sources (i.e., TV networks, special interest channels, etc.)
schedule their air time with two types of information:
"programming" for the purpose of informing or entertaining, and
"avails" for the purpose of advertising. The avails may occupy
roughly 20-25% of the total transmitting time, and are usually
divided into smaller intervals of 15, 30, or 60 seconds.
[0005] In many prior art systems, the insertion of advertisements
in avails is handled by a combination of cue-tone detectors,
switching equipment and tape players that hold the advertising
material. Upon receipt of the cue tones, an insertion controller
automatically turns on a tape player containing the advertisement.
Switching equipment then switches the system output from the video
and audio signals received from the programming source to the
output of the tape player. The tape player remains on for the
duration of the advertising, after which the insertion controller
causes the switching equipment to switch back to the video and
audio channels of the programming source. When switched, these
successive program and advertising segments usually feed to a
radio-frequency (RF) modulator for delivery to the subscribers.
[0006] Many subscriber television systems, such as cable television
are currently being converted to digital systems. These new digital
systems compress the advertising data according to decompression
standards, such as a Motion Picture Experts Group (MPEG)
compression standard (currently MPEG-2 standard). The compressed
data is then stored as a digital file on a large disk drive (or
several drives). Upon receipt of the cue tone, the digital file is
spooled ("played") off of the drive.
[0007] The advertisement may be inserted into the digital MPEG
stream using digital video splicing techniques that include the
healing of the broken MPEG stream. Alternatively, the digital
advertisement may be converted to analog and spliced with an analog
signal. Yet another technique for ad insertion involves
decompressing the digital MPEG stream and splicing the ad in with
the program in an uncompressed format.
[0008] A prior art (present model) of providing advertisements
along with actual programming is based on linked sponsorship. In
the linked sponsorship model, the advertisements are inserted into
the actual programming based on the contents of the programming,
e.g., a baby stroller advertisement may be inserted into a
parenting program.
[0009] Even with linked sponsorship, advertising, and in particular
broadcast television advertising, is mostly ineffective. That is, a
large percentage, if not the majority of advertisements, do not
have a high probability of effecting a sale. In addition to this
fact, many advertisements are not even seen/heard by the subscriber
who may mute the sound, change channels, or simply leave the room
during a commercial break.
[0010] The reasons for such ineffectiveness are due to the fact
that the displayed advertisements are not targeted to the
subscribers' needs, likes or preferences. Generally, the same
advertisements are displayed to all the subscribers irrespective of
the needs and preferences of the subscribers.
[0011] In the Internet world, efforts have been made to collect
information about subscriber likes and preferences by different
means, e.g., by the use of cookies. In cookies and other profiling
means, the user viewing habits, purchase habits, or surfing habits
are monitored, recorded and analyzed, and then, based on the
analysis, suitable advertisements are selected. Even though cookies
and other profiling means assist in targeting advertising, they
have recently come under fire as these means are known to invade
the privacy of the subscribers without their authorization.
[0012] Thus, a system and a method is desired for providing
subscribers/consumers with advertisements which are more
targeted/directed to their lifestyles, while ensuring that their
demographic, purchase, and product preference data is maintained
private.
SUMMARY OF THE INVENTION
[0013] The present invention is directed at a system and a method
for providing subscribers/consumers with advertisements that are
more directed to their lifestyles, while ensuring that their
demographic, purchase, and product preference data is maintained
private. The present invention allows manufacturers and advertisers
to use their advertising dollars more effectively across a
multitude of media platforms including video and Internet domains,
and eventually extending into the printed media.
[0014] The system is based on the premise that the subscribers may
agree to have advertisements delivered to them on a more selective
basis than the prior art "linked sponsorship" model in which
advertisements are only linked to the contents of the programming.
Subscribers/consumers who sign up for this service will receive
discounts from the Internet access or video service provider.
Advertisers may send profiles for their advertisements to a Secure
Correlation Server.TM. (SCS) that allows the advertisement to be
correlated to the subscriber profiles. No information regarding the
subscriber is released, and subscribers who do not wish to
participate in the service are not profiled.
[0015] A system in accordance with one embodiment of the present
invention utilizes the principles of Quantum Advertising.TM. in
which subscribers/consumers are described by consumer/subscriber
characterization vectors that contain deterministic and
probabilistic information regarding the consumer/subscriber, but do
not contain privacy violating information such as, transaction
records of purchases, video selections, or other raw data.
[0016] In accordance with the principles of one embodiment of the
present invention, the subscriber profiles may be created by
collecting information from a plurality of distributed databases.
These distributed databases may be queried through the use of
operators that in effect make measurements on certain
"observables." By controlling the types of observables, certain
parameters may be measured (in a probabilistic or deterministic
sense) while other parameters may remain unmeasurable in order to
preserve privacy. The operators may include clustering operators as
well as operators for correlating advertisement characterization
vectors with consumer/subscriber characterizations.
[0017] In another embodiment of the present invention, a system
permits the targeting of advertisements in the Internet and video
platforms, e.g., Switched Digital Video (SDV) and cable-based
systems. In a SDV platform, the present invention allows for
resolution of the advertising at the level of the home and even at
the level of the individual user/subscriber. The system of the
present invention may also be utilized for the delivery of
advertisements over cable networks by selecting advertisements at
the head end or substituting advertisements in the set-top box.
[0018] The general principles of the present invention are not
constrained to video networks and may be generally applied to a
variety of media systems including printed media, radio
broadcasting, and store coupons. The system provides the overall
capability to match advertisements using consumer profiles that do
not contain the raw transaction information, thus subscriber
privacy is maintained.
BRIEF DESCRIPTION OF THE DRAWINGS
[0019] These and other features and objects of the invention will
be more fully understood from the following detailed description of
the preferred embodiments that should be read in light of the
accompanying drawings:
[0020] FIG. 1A illustrates advertisement applicability modeled as a
distribution curve;
[0021] FIG. 1B illustrates an exemplary case of targeted marketing,
where subscribers are divided into subgroups and the advertisement
is displayed only to a subgroup of the subscribers;
[0022] FIG. 1C illustrates an exemplary case where different
success rates are determined by measuring products or services that
were purchased as the result of the viewing of a targeted
advertisement;
[0023] FIG. 2 illustrates an exemplary television system based on
traditional advertising schemes;
[0024] FIG. 3 illustrates a system utilizing targeted
advertisements based on the principles of the present
invention;
[0025] FIG. 4 illustrates a context diagram for a subscriber
characterization system.
[0026] FIG. 5 illustrates a block diagram for a realization of a
subscriber monitoring system for receiving video signals;
[0027] FIG. 6 illustrates a block diagram of a channel
processor;
[0028] FIG. 7 illustrates a channel sequence and volume over a
twenty-four (24) hour period;
[0029] FIG. 8 illustrates a time of day detailed record;
[0030] FIG. 9 illustrates a household viewing habits statistical
table;
[0031] FIG. 10A illustrates an entity-relationship diagram for the
generation of program characteristics vectors;
[0032] FIG. 10B illustrates a flowchart for program
characterization;
[0033] FIG. 11A illustrates a deterministic program category
vector;
[0034] FIG. 11B illustrates a deterministic program sub-category
vector;
[0035] FIG. 11C illustrates a deterministic program rating
vector;
[0036] FIG. 11D illustrates a probabilistic program category
vector;
[0037] FIG. 11E illustrates a probabilistic program sub-category
vector;
[0038] FIG. 11F illustrates a probabilistic program content
vector;
[0039] FIG. 12A illustrates a set of logical heuristic rules;
[0040] FIG. 12B illustrates a set of heuristic rules expressed in
terms of conditional probabilities;
[0041] FIG. 13 illustrates an entity-relationship diagram for the
generation of program demographic vectors;
[0042] FIG. 14 illustrates a program demographic vector;
[0043] FIG. 15 illustrates an entity-relationship diagram for the
generation of household session demographic data and household
session interest profiles;
[0044] FIG. 16 illustrates an entity-relationship diagram for the
generation of average and session household demographic
characteristics;
[0045] FIG. 17 illustrates average and session household
demographic data;
[0046] FIG. 18 illustrates an entity-relationship diagram for
generation of a household interest profile;
[0047] FIG. 19 illustrates a household interest profile including
programming and product profiles;
[0048] FIGS. 20A-B illustrate user relationship diagrams for the
present invention;
[0049] FIGS. 21A-D illustrate a probabilistic consumer demographic
characterization vector, a deterministic consumer demographic
characterization vector, a consumer product preference
characterization vector, and a storage structure for consumer
characterization vectors respectively;
[0050] FIGS. 22A-B illustrate an advertisement demographic
characterization vector and an advertisement product preference
characterization vector respectively;
[0051] FIG. 23 illustrates a context diagram for the present
invention;
[0052] FIGS. 24A-B illustrate pseudocoele updating the
characteristics vectors and for a correlation operation
respectively;
[0053] FIG. 25 illustrates heuristic rules;
[0054] FIGS. 26A-B illustrate flowcharts for updating consumer
characterization vectors and a correlation operation
respectively;
[0055] FIG. 27 represents pricing as a function of correlation;
[0056] FIG. 28 illustrates a representation of a consumer
characterization as a set of basis vectors and an ad
characterization vector;
[0057] FIG. 29 illustrates an exemplary implementation of
distributed databases, each of which contain a portion of
information that can be utilized to create a subscriber/consumer
profile; and
[0058] FIGS. 30A-B illustrate examples of demographic factors
including household size and ethnicity.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT
[0059] In describing a preferred embodiment of the invention
illustrated in the drawings, specific terminology will be used for
the sake of clarity. However, the invention is not intended to be
limited to the specific terms so selected, and it is to be
understood that each specific term includes all technical
equivalents which operate in a similar manner to accomplish a
similar purpose.
[0060] With reference to the drawings, in general, and FIGS. 1A
through 30B in particular, the apparatus of the present invention
is disclosed.
[0061] The principles of the present invention propose a method and
system for targeting advertisements to only a selected group of
subscribers without jeopardizing the privacy of the subscribers. As
illustrated in FIG. 1A, advertisement applicability, in accordance
with the principles of the present invention may be modeled as a
distribution curve. As illustrated in FIG. 1A, a well-designed
advertisement may be found to be "applicable" by the majority of
subscribers, but there will be a number of subscribers for whom the
advertisement will not be applicable. Similarly, some of the
subscribers may find the advertisement to be quite applicable or
extremely applicable. The subscribers that find the advertisement
to be extremely applicable are most likely to purchase the product
or service, and the subscribers that find the advertisement to be
less applicable are less likely to purchase the product or
service.
[0062] Thus, in accordance with the principles of the present
invention, the overall subscribership may be divided into subgroups
(smaller groups), and the advertisement may be displayed only to
the subgroup that is most interested in the advertisement and is
most likely to purchase the product. FIG. 1B illustrates an
exemplary case where subscribers are divided into subgroups, and
the advertisement is displayed only to a subgroup of the
subscribers.
[0063] By forming subgroups and targeting advertisements to one or
more subgroups, the effectiveness of the advertisements may be
greatly increased, and overall advertisement success rates may be
increased. The increase in overall advertisement success rates
represents more effective use of advertising dollars, and is a
"welfare gain" in the sense that those dollars may be used for
other goods and services. FIG. 1C illustrates an exemplary case
where different success rates are determined by measuring products
or services that were purchased as the result of the viewing of an
advertisement. As can be seen, the highest success rate corresponds
to the subgroup that finds the advertisement to be extremely
applicable, and the lowest success rate corresponds to the subgroup
that finds the advertisement least applicable.
[0064] The principles of the present invention may be applied to
many different applications. In one embodiment, the present
invention is utilized in a cable-based television (CTV) system.
FIG. 2 illustrates an exemplary CTV system based on a traditional
advertising business model. The CTV system consists of a content
provider 203 (e.g., programmers) producing syndicated programs
having advertising spots (avails). The content provider 203 also
incorporates national advertisements that are received from a
national advertiser 205. The programming contents (along with
national advertisements) are then provided to a network operator
(e.g., cable operator) 207. Generally, the network operator 207
purchases the programming contents for a fee. The network provider
207 is also provided with a right to substitute a percentage of the
national advertisements with local advertisements (e.g. 20% of the
advertisements may be substituted).
[0065] Thus, the network operator 207 may directly receive from one
or more local advertisers 209 local advertisements to replace a
percentage of the national advertisements. The local advertisements
may also be received from the national advertiser 205. The network
operator 207 then delivers the advertisements and programming to
subscribers/consumer 215 via an access network 211. The information
may be delivered to a personal computer or a television or any
other display means at the subscriber end. The access network 211
may be a cable-based system, a satellite-based television system,
an Internet-based computer network, or a Switched Digital Video
(SDV) platform using xDSL transmission technology. Such access
systems are well known to those skilled in the art.
[0066] In traditional systems, e.g., in the exemplary system of
FIG. 2, the local advertisements are not generally customized based
on the needs/preferences of the subscribers. Instead, the local
advertisements are selected based on local markets, and the same
advertisement is displayed to a subgroup, e.g., the opening of a
local store may be advertised to a few local subscribers. Thus,
even though the traditional advertising scheme as illustrated in
FIG. 2 attempts to substitute national/generic advertisements with
some local advertisements, the effectiveness of the advertisements
is not increased because the advertisements are not
customized/tailored based on user preferences/likes.
[0067] FIG. 3 illustrates a system utilizing targeted
advertisements based on the principles of the present invention. In
this model, the local advertisements are delivered from the
advertisers to a centralized Secure Correlation Server.TM. 305
configured to perform matching of the advertisements to subscribers
or groups of subscribers. At the correlation server 305, the input
is received from a secure profiling system 307 in the form of
subscriber profiles, and advertisements are matched to one or more
subscriber profiles.
[0068] As illustrated in FIG. 3, a content provider 303 receives
national advertisements from one or more advertisers 301,
multiplexes the national advertisements in the programming and
forwards the program streams having national advertisements to the
correlation server 305. The correlation server 305 evaluates the
advertisements and attempts to match them with one or more
subscriber profiles stored in the secure profiling system 307. The
correlation server 305, based on one or more subscriber profiles
can substitute national advertisements within the program streams
with more targeted advertisements received from local advertisers
309 or from national advertisers 311. The correlation server 305
may also receive local advertisements from the advertisers 301.
[0069] The correlation server 305 forwards programming having
targeted advertisements to a network operator 313. The programming
having targeted advertisements may then be forwarded to a
subscriber 317 via an access network 315. On the subscriber end,
the information may be delivered to a personal computer or a
television or any other display means.
[0070] FIG. 3 illustrates the ability of a system in accordance
with the principles of the present invention to target national
advertisements as well as local advertisements. The advertisers may
provide national advertisements to a Secure Correlation Server.TM.
305 that may match the advertisements to different subscribers 317.
By providing the ability to match advertisements to demographic
groups (in cable television systems) and to individual subscribers
(in switched digital video systems) using the correlation process,
the present invention allows for substantial increases in
advertising effectiveness.
[0071] The system of FIG. 3 is secure for many reasons. First, the
correlation server 305 does not contain raw data such as viewing or
purchase records. Second, the correlation server 305 does not
transmit subscriber/consumer profiles to third parties, and only
performs internal calculations to determine the applicability of an
advertisement to an individual subscriber.
[0072] It is to be noted that even though previously described
embodiments are described with reference to video advertisements,
the principles of the present invention are not based on a
particular media. The principles of the present invention may be
applied to diverse media such as printed media in which there are
national (broadcast) advertisements as well as local
advertisements, Internet advertisements, radio advertisements (in
particular Internet radio broadcasting) and a variety of other
forms of media advertisements.
[0073] In accordance with the principles of the present invention,
different types of profiles may be created by the secure profiling
system 307. These profiles may be subscriber profiles created from
video selection data, consumer profiles created from retail
purchases, and profiles created from the voluntary information
provided by the consumer/subscriber. In a switched digital video
system, these profiles may be based on individual viewing habits.
In cable-based television systems, these profiles may be based on
specific pay-per-view demands. In Internet-based computer networks,
these profiles may be based upon Internet surfing habits.
[0074] As discussed above, one type of profile that can be
generated is based on video selection data. The programming viewed
by the subscriber, both entertainment and advertisement, can be
studied and processed by a subscriber characterization system to
determine program characteristics. This determination of the
program characteristics is referred to as a program characteristics
vector. The vector may be a truly one-dimensional vector, but can
also be represented as an n dimensional matrix that can be
decomposed into vectors.
[0075] The subscriber profile vector represents a profile of the
subscriber (or the household of subscribers) and can be in the form
of a demographic profile (average or session) or a program or
product preference vector. The program and product preference
vectors are considered to be part of a household interest profile
that can be thought of as an n dimensional matrix representing
probabilistic measurements of subscriber interests.
[0076] In the case that the subscriber profile vector is a
demographic profile, the subscriber profile vector indicates a
probabilistic measure of the age of the subscriber or average age
of the viewers in the household, sex of the subscriber, income
range of the subscriber or household, and other such demographic
data. Such information comprises household demographic
characteristics and is composed of both average and session values.
Extracting a single set of values from the household demographic
characteristics can correspond to a subscriber profile vector.
[0077] The household interest profile can contain both programming
and product profiles, with programming profiles corresponding to
probabilistic determinations of what programming the subscriber
(household) is likely to be interested in, and product profiles
corresponding to what products the subscriber (household) is likely
to be interested in. These profiles contain both an average value
and a session value, the average value being a time average of
data, where the averaging period may be several days, weeks,
months, or the time between resets of unit.
[0078] Since a viewing session is likely to be dominated by a
particular viewer, the session values may, in some circumstances,
correspond most closely to the subscriber values, while the average
values may, in some circumstances, correspond most closely to the
household values.
[0079] FIG. 4 depicts the context diagram of a preferred embodiment
of a Subscriber Characterization System (SCS) 400. A context
diagram, in combination with entity-relationship diagrams, provide
a basis from which one skilled in the art can realize the present
invention. The present invention can be realized in a number of
programming languages including C, C++, Perl, and Java, although
the scope of the invention is not limited by the choice of a
particular programming language or tool. Object oriented languages
have several advantages in terms of construction of the software
used to realize the present invention, although the present
invention can be realized in procedural or other types of
programming languages known to those skilled in the art.
[0080] In generating a subscriber profile, the SCS 400 receives
from a user 420 commands in the form of a volume control signal 424
or program selection data 422 which can be in the form of a channel
change but may also be an address request which requests the
delivery of programming from a network address. A record signal 426
indicates that the programming or the address of the programming is
being recorded by the user. The record signal 426 can also be a
printing command, a tape recording command, a bookmark command or
any other command intended to store the program being viewed, or
program address, for later use.
[0081] The material being viewed by the user 420 is referred to as
source material 430. The source material 430, as defined herein, is
the content that a subscriber selects and may consist of analog
video, Motion Picture Expert Group (MPEG) digital video source
material, other digital or analog material, Hypertext Markup
Language (HTML) or other type of multimedia source material. The
subscriber characterization system 400 can access the source
material 430 received by the user 420 using a start signal 432 and
a stop signal 434, which control the transfer of source related
text 436 which can be analyzed as described herein.
[0082] In a preferred embodiment, the source related text 436 can
be extracted from the source material 430 and stored in memory. The
source related text 436, as defined herein, includes source related
textual information including descriptive fields that are related
to the source material 430, or text that is part of the source
material 430 itself. The source related text 436 can be derived
from a number of sources including but not limited to closed
captioning information, Electronic Program Guide (EPG) material,
and text information in the source itself (e.g. text in HTML
files).
[0083] Electronic Program Guide (EPG) 440 contains information
related to the source material 430 that is useful to the user 420.
The EPG 440 is typically a navigational tool that contains source
related information including but not limited to the programming
category, program description, rating, actors, and duration. The
structure and content of EPG data is described in detail in U.S.
Pat. No. 5,596,373 assigned to Sony Corporation and Sony
Electronics that is herein incorporated by reference. As shown in
FIG. 4, the EPG 440 can be accessed by the SCS 400 by a request EPG
data signal 442 that results in the return of a category 444, a
sub-category 446, and a program description 448.
[0084] In one embodiment of the present invention, EPG data is
accessed and program information such as the category 444, the
sub-category 446, and the program description 448 are stored in
memory.
[0085] In another embodiment of the present invention, the source
related text 436 is the closed captioning text embedded in the
analog or digital video signal. Such closed captioning text can be
stored in memory for processing to extract the program
characteristic vectors 450.
[0086] One of the functions of the SCS 400 is to generate the
program characteristics vectors 450 which are comprised of program
characteristics data 452, as illustrated in FIG. 4. The program
characteristics data 452, which can be used to create the program
characteristics vectors 450 both in vector and table form, are
examples of source related information that represent
characteristics of the source material. In a preferred embodiment,
the program characteristics vectors 450 are lists of values that
characterize the programming (source) material in according to the
category 444, the sub-category 446, and the program description
448. The present invention may also be applied to advertisements,
in which case program characteristics vectors contain, as an
example, a product category, a product sub-category, and a brand
name.
[0087] As illustrated in FIG. 4, the SCS 400 uses heuristic rules
460. The heuristic rules 460, as described herein, are composed of
both logical heuristic rules as well as heuristic rules expressed
in terms of conditional probabilities. The heuristic rules 460 can
be accessed by the SCS 400 via a request rules signal 462 that
results in the transfer of a copy of rules 464 to the SCS 400.
[0088] The SCS 400 forms program demographic vectors 470 from
program demographics 472, as illustrated in FIG. 4. The program
demographic vectors 470 also represent characteristics of source
related information in the form of the intended or expected
demographics of the audience for which the source material is
intended.
[0089] Subscriber selection data 410 is obtained from the monitored
activities of the user and in a preferred embodiment can be stored
in a dedicated memory. In an alternate embodiment, the subscriber
selection data 410 is stored in a storage disk. Information that is
utilized to form the subscriber selection data 410 includes time
412, which corresponds to the time of an event, channel ID 414,
program ID 416, volume level 418, channel change record 419, and
program title 417. A detailed record of selection data is
illustrated in FIG. 8.
[0090] In a preferred embodiment, a household viewing habits 495
illustrated in FIG. 4 is computed from the subscriber selection
data 410. The SCS 400 transfers household viewing data 497 to form
household viewing habits 495. The household viewing data 497 is
derived from the subscriber selection data 410 by looking at
viewing habits at a particular time of day over an extended period
of time, usually several days or weeks, and making some
generalizations regarding the viewing habits during that time
period.
[0091] The program characteristics vector 450 is derived from the
source related text 436 and/or from the EPG 440 by applying
information retrieval techniques. The details of this process are
discussed in accordance with FIG. 10.
[0092] The program characteristics vector 450 is used in
combination with a set of the heuristic rules 460 to define a set
of the program demographic vectors 470 illustrated in FIG. 4
describing the audience the program is intended for.
[0093] One output of the SCS 400 is a household profile including
household demographic characteristics 490 and a household interest
profile 480. The household demographic characteristics 490
resulting from the transfer of household demographic data 492, and
the household interest profile 480, resulting from the transfer of
household interests data 482. Both the household demographics
characteristics 490 and the household interest profile 480 have a
session value and an average value, as will be discussed
herein.
[0094] The monitoring system depicted in FIG. 5 is responsible for
monitoring the subscriber activities, and can be used to realize
the SCS 400. In a preferred embodiment, the monitoring system of
FIG. 5 is located in a television set-top device or in the
television itself. In an alternate embodiment, the monitoring
system is part of a computer that receives programming from a
network.
[0095] In an application of the system for television services, an
input connector 520 accepts the video signal coming either from an
antenna, cable television input, or other network. The video signal
can be analog or Digital MPEG. Alternatively, the video source may
be a video stream or other multimedia stream from a communications
network including the Internet.
[0096] In the case of either analog or digital video, selected
fields are defined to carry EPG data or closed captioning text. For
analog video, the closed captioning text is embedded in the
vertical blanking interval (VBI). As described in U.S. Pat. No.
5,579,005, assigned to Scientific-Atlanta, Inc., the EPG
information can be carried in a dedicated channel or embedded in
the VBI. For digital video, the closed captioning text is carried
as video user bits in a user data field. The EPG data is
transmitted as ancillary data and is multiplexed at the transport
layer with the audio and video data.
[0097] Referring to FIG. 5, a system control unit 500 receives
commands from the user 520, decodes the command and forwards the
command to the destined module. In a preferred embodiment, the
commands are entered via a remote control to a remote receiver 505
or a set of selection buttons 507 available at the front panel of
the system control unit 500. In an alternate embodiment, the
commands are entered by the user 420 via a keyboard.
[0098] The system control unit 500 also contains a Central
Processing Unit (CPU) 503 for processing and supervising all of the
operations of the system control unit 500, a Read Only Memory (ROM)
502 containing the software and fixed data, a Random Access Memory
(RAM) 504 for storing data. CPU 503, RAM 504, ROM 502, and I/O
controller 501 are attached to a master bus 506. A power supply in
a form of battery can also be included in the system control unit
500 for backup in case of power outage.
[0099] An input/output (I/O) controller 501 interfaces the system
control unit 500 with external devices. In a preferred embodiment,
the I/O controller 501 interfaces to the remote receiver 505 and a
selection button such as the channel change button on a remote
control. In an alternate embodiment, it can accept input from a
keyboard or a mouse.
[0100] The program selection data 422 is forwarded to a channel
processor 510. The channel processor 510 tunes to a selected
channel and the media stream is decomposed into its basic
components: the video stream, the audio stream, and the data
stream. The video stream is directed to a video processor module
530 where it is decoded and further processed for display to the TV
screen. The audio stream is directed to an audio processor 540 for
decoding and output to the speakers.
[0101] The data stream can be EPG data, closed captioning text,
Extended Data Service (EDS) information, a combination of these, or
an alternate type of data. In the case of EDS the call sign,
program name and other useful data are provided. In a preferred
embodiment, the data stream is stored in a reserved location of the
RAM 504. In an alternate embodiment, a magnetic disk is used for
data storage. The system control unit 500 writes also in a
dedicated memory, which in a preferred embodiment is the RAM 504,
the selected channel, the time 412 of selection, the volume level
418 and the program ID 416 and the program title 417. Upon
receiving the program selection data 422, the new selected channel
is directed to the channel processor 510 and the system control
unit 500 writes to the dedicated memory the channel selection end
time and the program title 417 at the time 412 of channel change.
The system control unit 500 keeps track of the number of channel
changes occurring during the viewing time via the channel change
record 419. This data forms part of the subscriber selection data
410.
[0102] The volume control signal 424 is sent to the audio processor
540. In a preferred embodiment, the volume level 418 selected by
the user 420 corresponds to the listening volume. In an alternate
embodiment, the volume level 418 selected by the user 420
represents a volume level to another piece of equipment such as an
audio system (home theatre system) or to the television itself. In
such a case, the volume can be measured directly by a microphone or
other audio sensing device that can monitor the volume at which the
selected source material is being listened.
[0103] A program change occurring while watching a selected channel
is also logged by the system control unit 500. Monitoring the
content of the program at the time of the program change can be
done by reading the content of the EDS. The EDS contains
information such as program title, which is transmitted via the
VBI. A change on the program title field is detected by the
monitoring system and logged as an event. In an alternate
embodiment, an EPG is present and program information can be
extracted from the EPG. In a preferred embodiment, the programming
data received from the EDS or EPG permits distinguishing between
entertainment programming and advertisements.
[0104] FIG. 6 illustrates the block diagram of the channel
processor 510. In a preferred embodiment, the input connector 520
connects to a tuner 600 that tunes to the selected channel. A local
oscillator can be used to heterodyne the signal to the IF signal. A
demodulator 602 demodulates the received signal and the output is
fed to an FEC decoder 604. The data stream received from the FEC
decoder 604 is, in a preferred embodiment, in an MPEG format. In a
preferred embodiment, system demultiplexer 606 separates out video
and audio information for subsequent decompression and processing,
as well as ancillary data which can contain program related
information.
[0105] The data stream presented to the system demultiplexer 606
consists of packets of data including video, audio and ancillary
data. The system demultiplexer 606 identifies each packet from the
stream ID and directs the stream to the corresponding processor.
The video data is directed to the video processor module 530 and
the audio data is directed to the audio processor 540. The
ancillary data can contain closed captioning text, emergency
messages, program guide, or other useful information.
[0106] Closed captioning text is considered to be ancillary data
and is thus contained in the video stream. The system demultiplexer
606 accesses the user data field of the video stream to extract the
closed captioning text. The program guide, if present, is carried
on data stream identified by a specific transport program
identifier.
[0107] In an alternate embodiment, analog video can be used. For
analog programming, ancillary data such as closed captioning text
or EDS data are carried in a vertical blanking interval.
[0108] FIG. 7 illustrates a channel sequence and volume over a
twenty-four (24) hour period. The Y-axis represents the status of
the receiver in terms of on/off status and volume level. The X-axis
represents the time of day. The channels viewed are represented by
the windows 701-706, with a first channel 702 being watched
followed by the viewing of a second channel 704, and a third
channel 706 in the morning. In the evening a fourth channel 701 is
watched, a fifth channel 703, and a sixth channel 705. A channel
change is illustrated by a momentary transition to the "off" status
and a volume change is represented by a change of level on the
Y-axis.
[0109] A detailed record of the subscriber selection data 410 is
illustrated in FIG. 8 in a table format. A time column 802 contains
the starting time of every event occurring during the viewing time.
A Channel ID column 804 lists the channels viewed or visited during
that period. A program title column 803 contains the titles of all
programs viewed. A volume column 801 contains the volume level 418
at the time 412 of viewing a selected channel.
[0110] A representative statistical record corresponding to the
household viewing habits 495 is illustrated in FIG. 9. In a
preferred embodiment, a time of day column 900 is organized in
period of time including morning, mid-day, afternoon, night, and
late night. In an alternate embodiment, smaller time periods are
used. A minutes watched column 902 lists, for each period of time,
the time in minutes in which the SCS 400 recorded delivery of
programming. The number of channel changes during that period and
the average volume are also included in that table in a channel
changes column 904 and an average volume column 906 respectively.
The last row of the statistical record contains the totals for the
items listed in the minutes watched column 902, the channel changes
column 904 and the average volume 906.
[0111] FIG. 10A illustrates an entity-relationship diagram for the
generation of the program characteristics vector 450. The context
vector generation and retrieval technique described in U.S. Pat.
No. 5,619,709, which is incorporated herein by reference, can be
applied for the generation of the program characteristics vectors
450. Other techniques are well known by those skilled in the
art.
[0112] Referring to FIG. 10A, the source material 430 or the EPG
440 is passed through a program characterization process 1000 to
generate the program characteristics vectors 450. The program
characterization process 1000 is described in accordance with FIG.
10B. Program content descriptors including a first program content
descriptor 1002, a second program content descriptor 1004 and an
nth program content descriptor 1006, each classified in terms of
the category 444, the sub-category 446, and other divisions as
identified in the industry accepted program classification system,
are presented to a context vector generator 1020. As an example,
the program content descriptor can be text representative of the
expected content of material found in the particular program
category 444. In this example, the program content descriptors
1002, 1004 and 1006 would contain text representative of what would
be found in programs in the news, fiction, and advertising
categories respectively. The context vector generator 1020
generates context vectors for that set of sample texts resulting in
a first summary context vector 1008, a second summary context
vector 1010, and an nth summary context vector 1012. In the example
given, the summary context vectors 1008, 1010, and 1012 correspond
to the categories of news, fiction and advertising respectively.
The summary vectors are stored in a local data storage system.
[0113] Referring to FIG. 10B, a sample of the source related text
436 that is associated with the new program to be classified is
passed to the context vector generator 1020 that generates a
program context vector 1040 for that program. The source related
text 436 can be either the source material 430, the EPG 440, or
other text associated with the source material. A comparison is
made between the actual program context vectors and the stored
program content context vectors by computing, in a dot product
computation process 1030, the dot product of the first summary
context vector 1008 with the program context vector 1040 to produce
a first dot product 1014. Similar operations are performed to
produce second dot product 1016 and nth dot product 1018.
[0114] The values contained in the dot products 1014, 1016 and
1018, while not probabilistic in nature, can be expressed in
probabilistic terms using a simple transformation in which the
result represents a confidence level of assigning the corresponding
content to that program. The transformed values add up to one. The
dot products can be used to classify a program, or form a weighted
sum of classifications that results in the program characteristics
vectors 450. In the example given, if the source related text 436
was from an advertisement, the nth dot product 1018 would have a
high value, indicating that the advertising category was the most
appropriate category, and assigning a high probability value to
that category. If the dot products corresponding to the other
categories were significantly higher than zero, those categories
would be assigned a value, with the result being the program
characteristics vectors 450 as shown in FIG. 11D.
[0115] For the sub-categories, probabilities obtained from the
content pertaining to the same sub-category 446 are summed to form
the probability for the new program being in that sub-category 446.
At the sub-category level, the same method is applied to compute
the probability of a program being from the given category 444. The
three levels of the program classification system; the category
444, the sub-category 446 and the content, are used by the program
characterization process 1000 to form the program characteristics
vectors 450 which are depicted in FIGS. 11D-11F.
[0116] The program characteristics vectors 450 in general are
represented in FIGS. 11A-F. FIGS. 11A-C are an example of
deterministic program vectors. This set of vectors is generated
when the program characteristics are well defined, as can occur
when the source related text 436 or the EPG 440 contains specific
fields identifying the category 444 and the sub-category 446. A
program rating can also provided by the EPG 440.
[0117] In the case that these characteristics are not specified, a
statistical set of vectors is generated from the process described
in accordance with FIG. 10. FIG. 11D shows the probability that a
program being watched is from the given category 444. The
categories are listed in the X-axis. The sub-category 446 is also
expressed in terms of probability. This is shown in FIG. 11E. The
content component of this set of vectors is a third possible level
of the program classification, and is illustrated in FIG. 11F.
[0118] FIG. 12A illustrates sets of logical heuristics rules that
form part of the heuristic rules 460. In a preferred embodiment,
logical heuristic rules are obtained from sociological or
psychological studies. Two types of rules are illustrated in FIG.
12A. The first type links an individual's viewing characteristics
to demographic characteristics such as gender, age, and income
level. A channel changing rate rule 1230 attempts to determine
gender based on channel change rate. An income related channel
change rate rule 1210 attempts to link channel change rates to
income brackets. A second type of rules links particular programs
to particular audience, as illustrated by a gender determining rule
1250 which links the program category 444/sub-category 446 with a
gender. The result of the application of the logical heuristic
rules illustrated in FIG. 12A are probabilistic determinations of
factors including gender, age, and income level. Although a
specific set of logical heuristic rules has been used as an
example, a wide number of types of logical heuristic rules can be
used to realize the present invention. In addition, these rules can
be changed based on learning within the system or based on external
studies that provide more accurate rules.
[0119] FIG. 12B illustrates a set of the heuristic rules 460
expressed in terms of conditional probabilities. In the example
shown in FIG. 12B, the category 444 has associated with it
conditional probabilities for demographic factors such as age,
income, family size and gender composition. The category 444 has
associated with it conditional probabilities that represent
probability that the viewing group is within a certain age group
dependent on the probability that they are viewing a program in
that category 444.
[0120] FIG. 13 illustrates an entity-relationship diagram for the
generation of the program demographic vectors 470. In a preferred
embodiment, the heuristic rules 460 are applied along with the
program characteristic vectors 450 in a program target analysis
process 1300 to form the program demographic vectors 470. The
program characteristic vectors 450 indicate a particular aspect of
a program, such as its violence level. The heuristic rules 460
indicate that a particular demographic group has a preference for
that program. As an example, it may be the case that young males
have a higher preference for violent programs than other sectors of
the population. Thus, a program which has the program
characteristic vectors 450 indicating a high probability of having
violent content, when combined with the heuristic rules 460
indicating that "young males like violent programs", will result,
through the program target analysis process 1300, in the program
demographic vectors 470 which indicate that there is a high
probability that the program is being watched by a young male.
[0121] The program target analysis process 1300 can be realized
using software programmed in a variety of languages which processes
mathematically the heuristic rules 460 to derive the program
demographic vectors 470. The table representation of the heuristic
rules 460 illustrated in FIG. 12B expresses the probability that
the individual or household is from a specific demographic group
based on a program with a particular category 444. This can be
expressed, using probability terms as follow "the probability that
the individuals are in a given demographic group conditional to the
program being in a given category". Referring to FIG. 14, the
probability that the group has certain demographic characteristics
based on the program being in a specific category is
illustrated.
[0122] Expressing the probability that a program is destined to a
specific demographic group can be determined by applying Bayes
rule. This probability is the sum of the conditional probabilities
that the demographic group likes the program, conditional to the
category 444 weighted by the probability that the program is from
that category 444. In a preferred embodiment, the program target
analysis can calculate the program demographic vectors by
application of logical heuristic rules, as illustrated in FIG. 12A,
and by application of heuristic rules expressed as conditional
probabilities as shown in FIG. 12B. Logical heuristic rules can be
applied using logical programming and fuzzy logic using techniques
well understood by those skilled in the art, and are discussed in
the text by S. V. Kartalopoulos entitled "Understanding Neural
Networks and Fuzzy Logic" which is incorporated herein by
reference.
[0123] Conditional probabilities can be applied by simple
mathematical operations multiplying program context vectors by
matrices of conditional probabilities. By performing this process
over all the demographic groups, the program target analysis
process 1300 can measure how likely a program is to be of interest
to each demographic group. Those probabilities values form the
program demographic vector 470 represented in FIG. 14.
[0124] As an example, the heuristic rules expressed as conditional
probabilities shown in FIG. 12B are used as part of a matrix
multiplication in which the program characteristics vector 450 of
dimension N, such as those shown in FIGS. 11A-11F is multiplied by
an N.times.M matrix of heuristic rules expressed as conditional
probabilities, such as that shown in FIG. 12B. The resulting vector
of dimension M is a weighted average of the conditional
probabilities for each category and represents the household
demographic characteristics 490. Similar processing can be
performed at the sub-category and content levels.
[0125] FIG. 14 illustrates an example of the program demographic
vector 470, and shows the extent to which a particular program is
destined to a particular audience. This is measured in terms of
probability as depicted in FIG. 14. The Y-axis is the probability
of appealing to the demographic group identified on the X-axis.
[0126] FIG. 15 illustrates an entity-relationship diagram for the
generation of household session demographic data 1510 and household
session interest profile 1520. In a preferred embodiment, the
subscriber selection data 410 is used along with the program
characteristics vectors 450 in a session characterization process
1500 to generate the household session interest profile 1520. The
subscriber selection data 410 indicates what the subscriber is
watching, for how long and at what volume they are watching the
program.
[0127] In a preferred embodiment, the session characterization
process 1500 forms a weighted average of the program
characteristics vectors 450 in which the time duration the program
is watched is normalized to the session time (typically defined as
the time from which the unit was turned on to the present). The
program characteristics vectors 450 are multiplied by the
normalized time duration (which is less than one unless only one
program has been viewed) and summed with the previous value. Time
duration data, along with other subscriber viewing information, is
available from the subscriber selection data 410. The resulting
weighted average of program characteristics vectors forms the
household session interest profile 1520, with each program
contributing to the household session interest profile 1520
according to how long it was watched. The household session
interest profile 1520 is normalized to produce probabilistic values
of the household programming interests during that session.
[0128] In an alternate embodiment, the heuristic rules 460 are
applied to both the subscriber selection data 410 and the program
characteristics vectors 450 to generate the household session
demographic data 1510 and the household session interest profile
1520. In this embodiment, weighted averages of the program
characteristics vectors 450 are formed based on the subscriber
selection data 410, and the heuristic rules 460 are applied. In the
case of logical heuristic rules as shown in FIG. 12A, logical
programming can be applied to make determinations regarding the
household session demographic data 1510 and the household session
interest profile 1520. In the case of heuristic rules in the form
of conditional probabilities such as those illustrated in FIG. 12B,
a dot product of the time averaged values of the program
characteristics vectors can be taken with the appropriate matrix of
heuristic rules to generate both the household session demographic
data 1510 and the household session interest profile 1520.
[0129] Volume control measurements which form part of the
subscriber selection data 410 can also be applied in the session
characterization process 1500 to form a household session interest
profile 1520. This can be accomplished by using normalized volume
measurements in a weighted average manner similar to how time
duration is used. Thus, muting a show results in a zero value for
volume, and the program characteristics vector 450 for this show
will not be averaged into the household session interest profile
1520.
[0130] FIG. 16 illustrates an entity-relationship diagram for the
generation of average household demographic characteristics and
session household demographic characteristics 490. A household
demographic characterization process 1600 generates the household
demographic characteristics 490 represented in table format in FIG.
17. The household demographic characterization process 1600 uses
the household viewing habits 495 in combination with the heuristic
rules 460 to determine demographic data. For example, a household
with a number of minutes watched of zero during the day may
indicate a household with two working adults. Both logical
heuristic rules as well as rules based on conditional probabilities
can be applied to the household viewing habits 495 to obtain the
household demographics characteristics 490.
[0131] The household viewing habits 495 is also used by the system
to detect out-of-habits events. For example, if a household with a
zero value for the minutes watched column 902 at late night
presents a session value at that time via the household session
demographic data 1510, this session will be characterized as an
out-of-habits event and the system can exclude such data from the
average if it is highly probable that the demographics for that
session are greatly different than the average demographics for the
household. Nevertheless, the results of the application of the
household demographic characterization process 1600 to the
household session demographic data 1510 can result in valuable
session demographic data, even if such data is not added to the
average demographic characterization of the household.
[0132] FIG. 17 illustrates the average and session household
demographic characteristics. A household demographic parameters
column 1701 is followed by an average value column 1705, a session
value column 1703, and an update column 1707. The average value
column 1705 and the session value column 1703 are derived from the
household demographic characterization process 1600. The
deterministic parameters such as address and telephone numbers can
be obtained from an outside source or can be loaded into the system
by the subscriber or a network operator at the time of
installation. Updating of deterministic values is prevented by
indicating that these values should not be updated in the update
column 1707.
[0133] FIG. 18 illustrates an entity-relationship diagram for the
generation of the household interest profile 480 in a household
interest profile generation process 1800. In a preferred
embodiment, the household interest profile generation process
comprises averaging the household session interest profile 1520
over multiple sessions and applying the household viewing habits
495 in combination with the heuristic rules 460 to form the
household interest profile 480 which takes into account both the
viewing preferences of the household as well as assumptions about
households/subscribers with those viewing habits and program
preferences.
[0134] FIG. 19 illustrates the household interest profile 480 that
is composed of a programming types row 1909, a products types row
1907, and a household interests column 1901, an average value
column 1903, and a session value column 1905.
[0135] The product types row 1907 gives an indication as to what
type of advertisement the household would be interested in
watching, thus indicating what types of products could potentially
be advertised with a high probability of the advertisement being
watched in its entirety. The programming types row 1909 suggests
what kind of programming the household is likely to be interested
in watching. The household interests column 1901 specifies the
types of programming and products that are statistically
characterized for that household.
[0136] As an example of the industrial applicability of the
invention, a household will perform its normal viewing routine
without being requested to answer specific questions regarding
likes and dislikes. Children may watch television in the morning in
the household, and may change channels during commercials, or not
at all. The television may remain off during the working day, while
the children are at school and day care, and be turned on again in
the evening, at which time the parents may "surf" channels, mute
the television during commercials, and ultimately watch one or two
hours of broadcast programming. The present invention provides the
ability to characterize the household, and may make the
determination that there are children and adults in the household,
with program and product interests indicated in the household
interest profile 480 corresponding to a family of that composition.
A household with two retired adults will have a completely
different characterization that will be indicated in the household
interest profile 480.
[0137] As discussed above, an additional method of profiling
includes profiling based on consumer purchases. FIG. 20A shows a
user relationship diagram that illustrates the relationships
between a consumer profiling system and various entities. As can be
seen in FIG. 20A, a consumer 2000 can receive information and
advertisements from a consumer personal computer (PC) 2004,
displayed on a television 2008 which is connected to a set-top
2006, or can receive a mailed ad 2082.
[0138] Advertisements and information displayed on consumer PC 2004
or television 2008 can be received over an Internet 2050, or can be
received over the combination of the Internet 2050 with another
telecommunications access system. The telecommunications access
system can include but is not limited to cable TV delivery systems,
switched digital video access systems operating over telephone
wires, microwave telecommunications systems, or any other medium
which provides connectivity between the consumer 2000 and a content
server 2062 and ad server 2046.
[0139] A content/opportunity provider 2060 maintains the content
server 2062 which can transmit content including broadcast
programming across a network such as the Internet 2050. Other
methods of data transport can be used including private data
networks and can connect the content sever 2060 through an access
system to a device owned by consumer 2000.
[0140] Content/opportunity provider 2060 is termed such since if
consumer 2000 is receiving a transmission from content server 2062,
the content/opportunity provider can insert an advertisement. For
video programming, content/opportunity provider is typically the
cable network operator or the source of entertainment material, and
the opportunity is the ability to transmit an advertisement during
a commercial break.
[0141] The majority of content that is being transmitted today is
done so in broadcast form, such as broadcast television programming
(broadcast over the air and via cable TV networks), broadcast
radio, and newspapers. Although the interconnectivity provided by
the Internet will allow consumer specific programming to be
transmitted, there will still be a large amount of broadcast
material that can be sponsored in part by advertising. The ability
to insert an advertisement in a broadcast stream (video, audio, or
mailed) is an opportunity for advertiser 2044. Content can also be
broadcast over the Internet and combined with existing video
services, in which case opportunities for the insertion of
advertisements will be present.
[0142] Although FIG. 20A represents content/opportunity provider
2060 and content server 2062 as being independently connected to
Internet 2050, with the consumer's devices being also being
directly connected to the Internet 2050, the content/opportunity
provider 2060 can also control access to the subscriber. This can
occur when the content/opportunity provider is also the cable
operator or telephone company. In such instances, the cable
operator or telephone company can be providing content to consumer
2000 over the cable operator/telephone company access network. As
an example, if the cable operator has control over the content
being transmitted to the consumer 2000, and has programmed times
for the insertion of advertisements, the cable operator is
considered to be a content/opportunity provider 2060 since the
cable operator can provide advertisers the opportunity to access
consumer 2000 by inserting an advertisement at the commercial
break.
[0143] In a preferred embodiment of the present invention, a
pricing policy can be defined. The content/opportunity provider
2060 can charge advertiser 2044 for access to consumer 2000 during
an opportunity. In a preferred embodiment the price charged for
access to consumer 2000 by content/opportunity provider varies as a
function of the applicability of the advertisement to consumer
2000. In an alternate embodiment consumer 2000 retains control of
access to the profile and charges for viewing an advertisement.
[0144] The content provider can also be a mailing company or
printer that is preparing printed information for consumer 2000. As
an example, content server 2062 can be connected to a printer 2064
that creates a mailed ad 2082 for consumer 2000. Alternatively,
printer 2064 can produce advertisements for insertion into
newspapers that are delivered to consumer 2000. Other printed
material can be generated by printer 2062 and delivered to consumer
2000 in a variety of ways.
[0145] Advertiser 2044 maintains an ad server 2046 that contains a
variety of advertisements in the form of still video that can be
printed, video advertisements, audio advertisements, or
combinations thereof.
[0146] Profiler 2040 maintains a consumer profile server 2030 that
contains the characterization of consumer 2000. The consumer
profiling system is operated by profiler 2040, who can use consumer
profile server 2030 or another computing device connected to
consumer profile server 2030 to profile consumer 2000.
[0147] Data to perform the consumer profiling is received from a
point of purchase 2010. Point of purchase 2010 can be a grocery
store, department store, other retail outlet, or can be a web site
or other location where a purchase request is received and
processed. In a preferred embodiment, data from the point of
purchase is transferred over a public or private network 2020, such
as a local area network within a store or a wide area network that
connects a number of department or grocery stores. In an alternate
embodiment the data from point of purchase 2010 is transmitted over
the Internet 2050 to profiler 2040.
[0148] Profiler 2040 may be a retailer who collects data from its
stores, but can also be a third party who contracts with consumer
2000 and the retailer to receive point of purchase data and profile
consumer 2000. Consumer 2000 may agree to such an arrangement based
on the increased convenience offered by targeted ads, or through a
compensation arrangement in which they are paid on a periodic basis
for revealing their specific purchase records.
[0149] FIG. 20B illustrates an alternate embodiment of the present
invention in which the consumer 2000 is also profiler 2040.
Consumer 2000 maintains consumer profile server 2030 that is
connected to a network, either directly or through consumer PC 2004
or set-top 2006. Consumer profile server 2030 can contain the
consumer profiling system, or the profiling can be performed in
conjunction with consumer PC 2004 or set-top 2006. A subscriber
characterization system that monitors the viewing habits of
consumer 2000 can be used in conjunction with the consumer
profiling system to create a more accurate consumer profile.
[0150] When the consumer 2000 is also the profiler 2040, as shown
in FIG. 20B, access to the consumer demographic and product
preference characterization is controlled exclusively by consumer
2000, who will grant access to the profile in return for receiving
an increased accuracy of ads, for cash compensation, or in return
for discounts or coupons on goods and services.
[0151] FIG. 21A illustrates an example of a probabilistic
demographic characterization vector. The demographic
characterization vector is a representation of the probability that
a consumer falls within a certain demographic category such as an
age group, gender, household size, or income range.
[0152] In a preferred embodiment the demographic characterization
vector includes interest categories. The interest categories may be
organized according to broad areas such as music, travel, and
restaurants. Examples of music interest categories include country
music, rock, classical, and folk. Examples of travel categories
include "travels to another state more than twice a year," and
travels by plane more than twice a year."
[0153] FIG. 21B illustrates a deterministic demographic
characterization vector. The deterministic demographic
characterization vector is a representation of the consumer profile
as determined from deterministic rather than probabilistic data. As
an example, if consumer 2000 agrees to answer specific questions
regarding age, gender, household size, income, and interests the
data contained in the consumer characterization vector will be
deterministic.
[0154] As with probabilistic demographic characterization vectors,
the deterministic demographic characterization vector can include
interest categories. In a preferred embodiment, consumer 2000
answers specific questions in a survey generated by profiler 2040
and administered over the phone, in written form, or via the
Internet 2050 and consumer PC 2004. The survey questions correspond
either directly to the elements in the probabilistic demographic
characterization vector, or can be processed to obtain the
deterministic results for storage in the demographic
characterization vector.
[0155] FIG. 21C illustrates a product preference vector. The
product preference represents the average of the consumer
preferences over past purchases. As an example, a consumer who buys
the breakfast cereal manufactured by Post under the trademark
ALPHABITS about twice as often as purchasing the breakfast cereal
manufactured by Kellogg under the trademark CORN FLAKES, but who
never purchases breakfast cereal manufactured by General Mills
under the trademark WHEATIES, would have a product preference
characterization such as that illustrated in FIG. 21C. As
illustrated in FIG. 21C, the preferred size of the consumer
purchase of a particular product type can also be represented in
the product preference vector.
[0156] FIG. 21D represents a data structure for storing the
consumer profile, which can be comprised of a consumer ID field
2137, a deterministic demographic data field 2139, a probabilistic
demographic data field 2141, and one or more product preference
data fields 2143. As illustrated in FIG. 21D, the product
preference data field 2143 can be comprised of multiple fields
arranged by product categories 2153.
[0157] Depending on the data structure used to store the
information contained in the vector, any of the previously
mentioned vectors may be in the foam of a table, record, linked
tables in a relational database, series of records, or a software
object.
[0158] A consumer ID 2312 (described later with respect to FIG. 23)
can be any identification value uniquely associated with consumer
2000. In a preferred embodiment consumer ID 2312 is a telephone
number, while in an alternate embodiment consumer ID 2312 is a
credit card number. Other unique identifiers include consumer name
with middle initial or a unique alphanumeric sequence, the consumer
address, social security number.
[0159] The vectors described and represented in FIGS. 21A-C form
consumer characterization vectors that can be of varying length and
dimension, and portions of the characterization vector can be used
individually. Vectors can also be concatenated or summed to produce
longer vectors that provide a more detailed profile of consumer
2000. A matrix representation of the vectors can be used, in which
specific elements, such a product categories 2153, are indexed.
Hierarchical structures can be employed to organize the vectors and
to allow hierarchical search algorithms to be used to locate
specific portions of vectors.
[0160] FIGS. 22A-B represent an ad demographics vector and an ad
product preference vector respectively. The ad demographics vector,
similar in structure to the demographic characterization vector, is
used to target the ad by setting the demographic parameters in the
ad demographics vector to correspond to the targeted demographic
group. As an example, if an advertisement is developed for a market
which is the 18-24 and 24-32 age brackets, no gender bias, with a
typical household size of 2-5, and income typically in the range of
$20,000-$50,000, the ad demographics vector would resemble the one
shown in FIG. 22A. The ad demographics vector represents a
statistical estimate of who the ad is intended for, based on the
advertisers belief that the ad will be beneficial to the
manufacturer when viewed by individuals in those groups. The
benefit will typically be in the form of increased sales of a
product or increased brand recognition. As an example, an "image
ad" which simply shows an artistic composition but which does not
directly sell a product may be very effective for young people, but
may be annoying to older individuals. The ad demographics vector
can be used to establish the criteria that will direct the ad to
the demographic group of 18-24 year olds.
[0161] FIG. 22B illustrates an ad product preference vector. The ad
product preference vector is used to select consumers that have a
particular product preference. In the example illustrated in FIG.
22B, the ad product preference vector is set so that the ad can be
directed a purchasers of ALPHABITS and WHEATIES, but not at
purchasers of CORN FLAKES. This particular setting would be useful
when the advertiser represents Kellogg and is charged with
increasing sales of CORN FLAKES. By targeting present purchasers of
ALPHABITS and WHEATIES, the advertiser can attempt to sway those
purchasers over to the Kellogg brand and in particular convince
them to purchase CORN FLAKES. Given that there will be a payment
required to present the advertisement, in the form of a payment to
the content/opportunity provider 2060 or to the consumer 2000, the
advertiser 2044 desires to target the ad and thereby increase its
cost effectiveness.
[0162] In the event that advertiser 2044 wants to reach only the
purchasers of Kellogg's CORN FLAKES, that category would be set at
a high value, and in the example shown would be set to 1. As shown
in FIG. 22B, product size can also be specified. If there is no
preference to size category the values can all be set to be equal.
In a preferred embodiment the values of each characteristic
including brand and size are individually normalized.
[0163] Because advertisements can be targeted based on a set of
demographic and product preference considerations which may not be
representative of any particular group of present consumers of the
product, the ad characterization vector can be set to identify a
number of demographic groups which would normally be considered to
be uncorrelated. Because the ad characterization vector can have
target profiles which are not representative of actual consumers of
the product, the ad characterization vector can be considered to
have discretionary elements. When used herein the term
discretionary refers to a selection of target market
characteristics which need not be representative of an actual
existing market or single purchasing segment.
[0164] In a preferred embodiment the consumer characterization
vectors shown in FIGS. 21A-C and the ad characterization vectors
represented in FIGS. 22A-B have a standardized format, in which
each demographic characteristic and product preference is
identified by an indexed position. In a preferred embodiment the
vectors are singly indexed and thus represent coordinates in
n-dimensional space, with each dimension representing a demographic
or product preference characteristic. In this embodiment a single
value represents one probabilistic or deterministic value (e.g. the
probability that the consumer is in the 18-24 year old age group,
or the weighting of an advertisement to the age group).
[0165] In an alternate embodiment a group of demographic or product
characteristics forms an individual vector. As an example, age
categories can be considered a vector, with each component of the
vector representing the probability that the consumer is in that
age group. In this embodiment each vector can be considered to be a
basis vector for the description of the consumer or the target ad.
The consumer or ad characterization is comprised of a finite set of
vectors in a vector space that describes the consumer or
advertisement.
[0166] FIG. 23 shows a context diagram for the present invention.
Context diagrams are useful in illustrating the relationship
between a system and external entities. Context diagrams can be
especially useful in developing object oriented implementations of
a system, although use of a context diagram does not limit
implementation of the present invention to any particular
programming language. The present invention can be realized in a
variety of programming languages including but not limited to C,
C++, Smalltalk, Java, Perl, and can be developed as part of a
relational database. Other languages and data structures can be
utilized to realize the present invention and are known to those
skilled in the art.
[0167] Referring to FIG. 23, in a preferred embodiment consumer
profiling system 2300 is resident on consumer profile server 2030.
Point of purchase records 2310 are transmitted from point of
purchase 2010 and stored on consumer profile server 2030. Heuristic
rules 2330, pricing policy 2370, and consumer profile 2360 are
similarly stored on consumer profile server 130. In a preferred
embodiment advertisement records 2340 are stored on ad server 2046
and connectivity between advertisement records 2340 and consumer
profiling system 2300 is via the Internet or other network.
[0168] In an alternate embodiment the entities represented in FIG.
23 are located on servers that are interconnected via the Internet
or other network.
[0169] Consumer profiling system 2300 receives purchase information
from a point of purchase, as represented by point of purchase
records 2310. The information contained within the point of
purchase records 2310 includes the consumer ID 2312, a product ID
2314 of the purchased product, the quantity 2316 purchased and the
price 2318 of the product. In a preferred embodiment, the date and
time of purchase 2320 are transmitted by point of purchase records
2310 to consumer profiling system 2300.
[0170] The consumer profiling system 2300 can access the consumer
profile 2360 to update the profiles contained in it. Consumer
profiling system 2300 retrieves a consumer characterization vector
2362 and a product preference vector 2364. Subsequent to retrieval
one or more data processing algorithms are applied to update the
vectors. An algorithm for updating is illustrated in the flowchart
in FIG. 26A. The updated vectors termed herein as new demographic
characterization vector 2366 and new product preference 2368 are
returned to consumer profile 2360 for storage.
[0171] Consumer profiling system 2300 can determine probabilistic
consumer demographic characteristics based on product purchases by
applying heuristic rules 2319. Consumer profiling system 2300
provides a product ID 2314 to heuristic rules records 2330 and
receives heuristic rules associated with that product. Examples of
heuristic rules are illustrated in FIG. 25.
[0172] In a preferred embodiment of the present invention, consumer
profiling system 2300 can determine the applicability of an
advertisement to the consumer 2000. For determination of the
applicability of an advertisement, a correlation request 2346 is
received by consumer profiling system 2300 from advertisements
records 2340, along with consumer ID 2312. Advertisements records
2340 also provides advertisement characteristics including an ad
demographic vector 2348, an ad product category 2352 and an ad
product preference vector 2354.
[0173] Application of a correlation process, as will be described
in accordance with FIG. 26B, results in a demographic correlation
2356 and a product correlation 2358 which can be returned to
advertisement records 2340. In a preferred embodiment, advertiser
2044 uses product correlation 2358 and demographic correlation 2356
to determine the applicability of the advertisement and to
determine if it is worth purchasing the opportunity. In a preferred
embodiment, pricing policy 2370 is utilized to determine an ad
price 2372 which can be transmitted from consumer profiling system
2300 to advertisement records 2340 for use by advertiser 2044.
[0174] Pricing policy 2370 is accessed by consumer profiling system
2300 to obtain ad price 2372. Pricing policy 2370 takes into
consideration results of the correlation provided by the consumer
profiling system 2300. An example of pricing schemes will be
discussed in detail later with respect to FIG. 27.
[0175] FIGS. 24A and 24B illustrate pseudocode for the updating
process and for a correlation operation respectively. The updating
process involves utilizing purchase information in conjunction with
heuristic rules to obtain a more accurate representation of
consumer 2000, stored in the form of a new demographic
characterization vector 2362 and a new product preference vector
2368.
[0176] As illustrated in the pseudocode in FIG. 24A the point of
purchase data are read and the products purchase are integrated
into the updating process. Consumer profiling system 2300 retrieves
a product demographics vector obtained from the set of heuristic
rules 2319 and applies the product demographics vector to the
demographics characterization vector 2362 and the product
preference vector 2364 from the consumer profile 2360.
[0177] The updating process as illustrated by the pseudocode in
FIG. 24A utilizes a weighting factor that determines the importance
of that product purchase with respect to all of the products
purchased in a particular product category. In a preferred
embodiment the weight is computed as the ratio of the total of
products with a particular product ID 2314 purchased at that time,
to the product total purchase, which is the total quantity of the
product identified by its product ID 2314 purchased by consumer
2000 identified by its consumer ID 2312, purchased over an extended
period of time. In a preferred embodiment the extended period of
time is one year.
[0178] In the preferred embodiment the product category total
purchase is determined from a record containing the number of times
that consumer 2000 has purchased a product identified by a
particular product ID.
[0179] In an alternate embodiment other types of weighting factors,
running averages and statistical filtering techniques can be used
to use the purchase data to update the demographic characterization
vector. The system can also be reset to clear previous demographic
characterization vectors and product preference vectors.
[0180] The new demographic characterization vector 2366 is obtained
as the weighted sum of the product demographics vector and the
demographic characterization vector 2362. The same procedure is
performed to obtain the new product preference vector 2368. Before
storing those new vectors, a normalization is performed on the said
new vectors. When used herein the term product characterization
information refers product demographics vectors, product purchase
vectors or heuristic rules, all of which can be used in the
updating process. The product purchase vector refers to the vector
that represents the purchase of an item represented by a product
ID. As an example, a product purchase vector for the purchase of
Kellogg's CORN FLAKES in a 32 oz. size has a product purchase
vector with a unity value for Kellogg's CORN FLAKES and in the 32
oz. size. In the updating process the weighted sum of the purchase
as represented by the product purchase vector is added to the
product preference vector to update the product preference vector,
increasing the estimated probability that the consumer will
purchase Kellogg's CORN FLAKES in the 32 oz. size.
[0181] In FIG. 24B the pseudocode for a correlation process is
illustrated. Consumer profiling system 2300, after receiving the
product characteristics and the consumer ID 2312 from the
advertisement records retrieves the consumer demographic
characterization vector 2362 and its product preference vector
2364. The demographic correlation is the correlation between the
demographic characterization vector 2362 and the ad demographics
vector. The product correlation is the correlation between the ad
product preference vector 2354 and the product preference vector
2364.
[0182] In a preferred embodiment the correlation process involves
computing the dot product between vectors. The resulting scalar is
the correlation between the two vectors.
[0183] In an alternate embodiment, as illustrated in FIG. 28, the
basis vectors which describe aspects of the consumer can be used to
calculate the projections of the ad vector on those basis vectors.
In this embodiment, the result of the ad correlation can itself be
in vector form whose components represent the degree of correlation
of the advertisement with each consumer demographic or product
preference feature. As shown in FIG. 28 the basis vectors are the
age of the consumer 2821, the income of the consumer 2801, and the
family size of the consumer 2831. The ad characterization vector
2850 represents the desired characteristics of the target audience,
and can include product preference as well as demographic
characteristics.
[0184] In this embodiment the degree of orthogonality of the basis
vectors will determine the uniqueness of the answer. The
projections on the basis vectors form a set of data that represent
the corresponding values for the parameter measured in the basis
vector. As an example, if household income is one basis vector, the
projection of the ad characterization vector on the household
income basis vector will return a result indicative of the target
household income for that advertisement.
[0185] Because basis vectors cannot be readily created from some
product preference categories (e.g. cereal preferences) an
alternate representation to that illustrated in FIG. 21C can be
utilized in which the product preference vector represents the
statistical average of purchases of cereal in increasing size
containers. This vector can be interpreted as an average measure of
the cereal purchased by the consumer in a given time period.
[0186] The individual measurements of correlation as represented by
the correlation vector can be utilized in determining the
applicability of the advertisement to the subscriber, or a sum of
correlations can be generated to represent the overall
applicability of the advertisement.
[0187] In a preferred embodiment individual measurements of the
correlations, or projections of the ad characteristics vector on
the consumer basis vectors, are not made available to protect
consumer privacy, and only the absolute sum is reported. In
geometric terms this can be interpreted as disclosure of the sum of
the lengths of the projections rather than the actual projections
themselves.
[0188] In an alternate embodiment the demographic and product
preference parameters are grouped to form sets of paired scores in
which elements in the consumer characterization vector are paired
with corresponding elements of the ad characteristics vector. A
correlation coefficient such as the Pearson product-moment
correlation can be calculated. Other methods for correlation can be
employed and are well known to those skilled in the art.
[0189] When the consumer characterization vector and the ad
characterization vector are not in a standardized format, a
transformation can be performed to standardize the order of the
demographic and product preferences, or the data can be decomposed
into sets of basis vectors which indicate particular attributes
such as age, income or family size.
[0190] FIG. 25 illustrates an example of heuristic rules including
rules for defining a product demographics vector. From the product
characteristics, a probabilistic determination of household
demographics can be generated. Similarly, the monthly quantity
purchased can be used to estimate household size. The heuristic
rules illustrated in FIG. 25 serve as an example of the types of
heuristic rules that can be employed to better characterize
consumer 2000 as a result of their purchases. The heuristic rules
can include any set of logic tests, statistical estimates, or
market studies that provide the basis for better estimating the
demographics of consumer 2000 based on their purchases.
[0191] In FIG. 26A the flowchart for updating the consumer
characterization vectors is depicted. The system receives data from
the point of purchase at receive point of purchase information step
2600. The system performs a test to determine if a deterministic
demographic characterization vector is available at deterministic
demographic information available step 2610 and, if not, proceeds
to update the demographic characteristics.
[0192] Referring to FIG. 26A, at read purchase ID info step 2620,
the product ID 2314 is read, and at update consumer demographic
characterization vector step 2630, an algorithm such as that
represented in FIG. 24A is applied to obtain a new demographic
characterization vector 2366, which is stored in the consumer
profile 2360 at store updated demographic characterization vector
step 2640.
[0193] The end test step 2650 can loop back to the read purchase ID
info 2620 if all the purchased products are not yet processed for
updating, or continue to the branch for updating the product
preference vector 2364. In this branch, the purchased product is
identified at read purchase ID info step 2620. An algorithm, such
as that illustrated in FIG. 24A for updating the product preference
vector 2364, is applied in update product preference vector step
2670. The updated vector is stored in consumer profile 2360 at
store product preference vector step 2680. This process is carried
out until all the purchased items are integrated in the updating
process.
[0194] FIG. 26B shows a flowchart for the correlation process. At
step 2700 the advertisement characteristics described earlier in
accordance with FIG. 23 along with the consumer ID are received by
consumer profiling system 2300. At step 2710 the demographic
correlation 2356 is computed and at step 2720 the product
preference correlation 2358 is computed. An illustrative example of
an algorithm for correlation is presented in FIG. 24b. The system
returns demographic correlation 2356 and product preference
correlation 2358 to the advertisement records 2340 before exiting
the procedure at end step 2750.
[0195] FIG. 27 illustrates two pricing schemes, one for
content/opportunity provider 160 based pricing 2770, which shows
increasing cost as a function of correlation. In this pricing
scheme, the higher the correlation, the more the
content/opportunity provider 2060 charges to air the
advertisement.
[0196] FIG. 27 also illustrates consumer based pricing 2760, which
allows a consumer to charge less to receive advertisements which
are more highly correlated with their demographics and
interests.
[0197] As an example of the industrial applicability of the
invention, a consumer 2000 can purchase items in a grocery store
that also acts as a profiler 2040 using a consumer profiling system
2300. The purchase record is used by the profiler to update the
probabilistic representation of customer 2000, both in terms of
their demographics as well as their product preferences. For each
item purchased by consumer 2000, product characterization
information in the form of a product demographics vector and a
product purchase vector is used to update the demographic
characterization vector and the product preference vector for
consumer 2000.
[0198] A content/opportunity provider 2060 may subsequently
determine that there is an opportunity to present an advertisement
to consumer 2000. Content/opportunity provider 2060 can announce
this opportunity to advertiser 2044 by transmitting the details
regarding the opportunity and the consumer ID 2312. Advertiser 2044
can then query profiler 2040 by transmitting consumer ID 2312 along
with advertisement specific information including the correlation
request 2346 and ad demographics vector 2348. The consumer
profiling system 2300 performs a correlation and determines the
extent to which the ad target market is correlated with the
estimated demographics and product preferences of consumer 2000.
Based on this determination advertiser 2044 can decide whether to
purchase the opportunity or not.
[0199] The principles of the present invention also provide novel
ways of collecting subscriber information, e.g., subscribers have
options to control the flow of information. In one implementation,
the subscribers decide whether they want to be enrolled in the
profiling, i.e., whether they want their viewing habits and other
information to be collected.
[0200] In this implementation, the data is collected with the
explicit permission of the consumer/subscriber, who enrolls in the
service and agrees to be profiled, similar to an "opt-in" feature.
In the "opt-in" feature, the subscriber/consumer is specifically
inquired whether he or she wants to be profiled. In exchange for
opt-in, the subscribers may receive economic benefit from the
service through discounts on cable service, discounts through
retail outlets, rebates from specific manufacturers, and other
incentive plans.
[0201] In the case of video services, the subscribers may be
presented with a series of enrollment screens that confirm the
subscribers' opt-in and ask the subscriber for specific demographic
information that may be used to create one or more subscriber
profiles.
[0202] In performing the enrollment process, it is possible to
obtain specific demographic information including household income,
size, and age distribution. Although this information is not
necessary for profiling, obtaining it from the subscriber allows
deterministic information to be used in conjunction with the
probabilistic information.
[0203] Other opt-in methods may be used for the different media. In
an Internet environment, a free browser add-on/plug-in may be used
that performs profiling through one or more secured techniques that
remove cookies, alters/hides surf streams. In this case, the
subscriber will have an option to enroll in a secure system that
permits profiling in a controlled and secure manner along with
providing economic incentives for participation in the profiling
process. Upon enrolling in the service, a profiling module may be
downloaded or activated that may perform the profiling through the
browser.
[0204] The principles of the present invention also support the
construction of distributed databases, each of which contain a
portion of information that is utilized to create a
subscriber/consumer profile. The distributed databases are
constructed such that no privacy violating information is contained
in one database, and the operators utilized to extract information
from each database preserve privacy and do not measure the
parameters that should not be observed.
[0205] In the actual formation of subscriber profiles, the system
may extract information from a plurality of databases and aggregate
portions of the information to create a subscriber profile. In the
aggregation of data, the emerging standards, such as XML, may be
used for the transport of the data and the standardized profiles
may be utilized to ensure that the secured server may effectively
combine the elements of the distributed profiling databases to
create a composite consumer/subscriber characterization vector
characterizing subscriber profiles.
[0206] As illustrated in FIG. 29, the distributed database may be
comprised of specific data sets including: purchase transaction
data 2901 obtained from a point-of-sale 2911 which may be a
physical point-of-sale or a virtual (Internet) point-of-sale;
Internet transaction data 2907 obtained from a PC 2917 or other
device connected to the Internet; video transaction data 2905
obtained in conjunction with a television/set-top combination 2925
or other video centric device; and demographic data 2903 obtained
from demographic data sources 2913. The examples of demographic
data sources include commercial databases such as the
MicroVision.TM. product from the Claritas Corporation.TM.. Other
public or private databases 2909 including those containing tax
information may also be used. Different distributed databases are
configured to a secure correlation server (SCS) 2915.
[0207] In the present invention, Quantum Advertising.TM. is
proposed wherein a probabilistic representation of an individual's
interests, in particular, products and services, is utilized and
specific private information about the individual is kept private.
In this way, it is possible for advertisers to effectively target
information to consumers without violating their privacy. The basis
for what is termed Quantum Advertising.TM. is derived from quantum
mechanics, and in particular rests on the concept that an
individual's information may be treated in a similar fashion to
electrons and other subatomic particles. In quantum mechanics, it
is possible to have a probabilistic representation of a particle,
but impossible to have a deterministic representation in which the
precise position of the particle is known.
[0208] In the present invention, the probabilistic descriptions of
subscribers along with a restricted set of operators are developed.
The restricted set of operators allows certain measurements to be
made, but prohibits privacy invading determinations. As an example,
an operator may be created and utilized that may indicate a
probability that an individual will potentially purchase a new
health care product, such as a shampoo or a toothpaste, but proper
construction of the database and operators would prohibit
determination of the individual's exact income in order to see if
they are a potential purchaser of that product.
[0209] Another example would be the development of a target group
for a new drug, such as an HIV related product. The proper
construction of the databases and operators may allow for the
formation of a group of individuals likely to be receptive to the
product, but would not allow identification of individuals in the
group, and the database would not contain health related
information such as HIV status.
[0210] Thus, the principles of the present invention utilize one or
more operators that allow the measurement of certain parameters
(non-deterministic parameters), but prohibit the measurement of
other parameters. In accordance with the principles of the present
invention, the description of an individual/household may be
contained in a vector which is described as the ket vector, using
the notation |A> where A represents the vector describing an
aspect of the individual/household.
[0211] The ket vector |A> can be described as the sum of
components such that
|A>=(a.sub.1.rho..sub.1+a.sub.2.rho..sub.2+ . . .
a.sub.n.rho..sub.n)+(b.sub.1.sigma..sub.1+b.sub.2.sigma..sub.2+ . .
. b.sub.n.sigma..sub.n)+(c.sub.1.tau..sub.1+c.sub.2.tau..sub.2+ . .
. c.sub.n.tau..sub.n)+(d.sub.1.nu..sub.1+d.sub.2.nu..sub.2+ . . .
d.sub.n.nu..sub.n)+(e.sub.1.omega..sub.1+e.sub.2.omega..sub.2+ . .
. e.sub.n.omega..sub.n)
[0212] where a.sub.n.rho..sub.n represents weighted demographic
factors that may be deterministic or probabilistic.
[0213] The other components of the ket vector |A> include:
[0214] b.sub.n.sigma..sub.n, which represents weighted
socio-economic factors;
[0215] c.sub.n.tau..sub.n, which represents weighted housing
factors;
[0216] d.sub.n.nu..sub.n, which represents weighted purchase
factors; and
[0217] e.sub.n.omega..sub.n, which represents weighted consumption
factors.
[0218] The elements of the ket vector |A> may be stored on
distributed databases, and the components within the groups above
can be mixed and stored in various locations. In addition, |A>
may not comprise all of the components listed above, but may
instead utilize only a subset of that information.
[0219] Consistent with the concepts of wave functions in quantum
mechanics, for each ket vector there is a corresponding bra vector
of the format <A|. In order to insure that the probabilities are
normalized, the identity <A|A>=1 is insured. Although the ket
and bra vectors are expected to be real entities, there is the
possibility of storing additional information in a complex ket
vector, in which case the corresponding bra vector will be
<A.sup.*|, and the normalization criteria is <A*|A>=1.
[0220] Having created the basic descriptions of the
households/individuals in the form of a distributed or centralized
database, a series of linear operations may be performed on the
database in order to obtain results that provide targeting
information. The linear operations may be performed using one or
more operators, which when applied to the database, yield a
measurable result. It is important to note that by proper
construction of the operators, it is possible to prevent
inappropriate (privacy violating) measurements from being made.
[0221] The generalized method for obtaining information from the
database is thus:
targeting information=<A|f|A>
[0222] where f is a single operation or series of operations that
result in a measurable quantity (observable). Through the
application of these operators it is possible to query the database
in a controlled manner and obtain information about a target group,
or to determine if an advertisement is applicable to an
individual/household (subscriber).
[0223] For determination of the applicability of an advertisement
to an individual/household, the advertiser can supply an ad
characterization vector along with the ID of an
individual/household, with the applicability of the advertisement
being determined as:
ad applicability=<A|AC{ID}|A>
[0224] where AC {ID} is the ad characterization vector that
contains an ID that may be at the individual, anonymous, or group
level. Examples of the possible IDs are as follows:
[0225] Individual Level: [0226] social security # [0227] address
[0228] credit card/courtesy card # [0229] phone #
[0230] Anonymous (e.g. Through the Use of Anonymous Transaction
Profiling): [0231] transaction ID (video transaction records)
[0232] transaction ID (purchase transaction records) [0233]
transaction ID (surfing transaction records)
[0234] Group: [0235] zip code [0236] area code/central office code
[0237] domain name [0238] employer
[0239] The use of individual/household IDs allows determination of
the applicability of an advertisement for a particular household or
individual. Anonymous transaction IDs may be used when no
information regarding the identity of the subscriber is being
provided, but when transaction profiles have been developed based
on the use of anonymous transaction profiling. Group IDs may be
utilized to determine applicability of an advertisement to a
particular group, with the basis for the grouping being geographic,
demographic, socio-economic, or through another grouping
mechanism.
[0240] The operators may result in a simple correlation operation
in which the operator contains an advertisement characterization
vector which is correlated against elements in the database, or may
be a series of operations which result in the determination of the
applicability of an advertisement, or determination of the product
preferences of a group or an individual.
[0241] The ad characterization vector contains a description of the
expected characteristics of the target market. The ad
characterization vector may be obtained from the advertiser, a
media buyer, or individual cognizant of the market to which the
advertisement is directed.
[0242] Other operators can be constructed so that functions other
than correlations can be performed. As an example, grouping or
clustering can be performed on the database by performing a series
of operations that identifies consumers with similar
characteristics. In addition to grouping or clustering, operators
can be constructed to identify a set of subscribers who are
candidates for a product based on specific selection criteria. As
an example, it is possible to construct an operator which returns a
list of subscribers likely to be interested in a product, with the
level of interest being determined from probabilistic elements such
as age, income, previous purchase profiles, Internet profiles, or
video selection profiles.
[0243] Proper construction of the database (and in particular
construction of the ket vectors and ket vector subcomponents) and
the operators ensures that privacy is maintained and prevents
direct reading of the data and inappropriate queries. Furthermore,
the actual transaction records (e.g. purchases, web surf streams,
or channels viewed) are never stored, and no privacy violating
information (e.g. medical conditions) are stored in the
database.
[0244] FIGS. 30A and 30B illustrate examples of demographic factors
including household size and ethnicity. As previously mentioned,
these functions may not necessarily be probabilistic, but may be
obtained from questionnaires presented to the subscriber that lead
to deterministic responses. These responses can be represented as
unity value probabilities.
[0245] The principles of the present invention propose advantages
both for subscribers and advertisers. A proposed service in
accordance with the principles of the present invention will be
free to the subscriber and includes incentives such as discounts on
Internet/video service. Furthermore, the advertisers may pay a
premium for advertisements placed using the system. This premium is
amongst the content provider, the Internet/video service provider,
and the provider of ad matching.
[0246] As an example, if an advertising opportunity during a
network sports event costs $0.10 per viewer, the charge for the
matched advertisement (ad) might be $0.14 per viewer. The
additional $0.04 is divided amongst the content provider (in this
case the network), the Internet/video service provider, and the
provider of ad matching. Because the ad is not displayed to the
entire set of viewers, but rather to the subset of viewers that
will find the ad acceptable, the total cost to the advertiser is
likely to be less than, or at most the same as without the
matching. The ad matching increases the effectiveness of the
advertising and thus makes better use of advertising dollars.
[0247] The service may be applied to cable networks, both for
Internet based services as well as video services. For Internet
based services over cable networks, the targeting may be at the
level of the individual home. For video services, the targeting is
presently at the level of the node, since cable networks do not
have the individual home resolution that switched digital video
networks have. Ad substitution technology at the set-top level may
increase the resolution of cable advertising, while SDV networks
are inherently capable of resolution at the individual home
level.
[0248] In one embodiment, the present invention may be deployed as
an Ad Management System (AMS) in a video environment. The AMS
includes a Secure Correlation Server.TM. (SCS) configured to
deliver targeted advertisements over video systems including
Switched Digital Video (SDV) platforms, cable platforms, satellite
platforms, and streaming video (Internet) delivery platforms. The
system allows for advertisers to deliver ad characterization
vectors to the Secure Correlation Server.TM. (SCS). The ad
characterization vectors assist in determining the applicability of
the advertisements to a particular subscriber or group of
subscribers (e.g., node). The AMS performs the functions of
prioritizing, selling, scheduling, and billing of video
advertisements.
[0249] In another embodiment, the present invention may be deployed
as a browser add-on/plug-in for the Internet environment. In this
embodiment, the profiling is not completely blocked, but the
subscriber is allowed to switch to a secured mode wherein the
subscriber is profiled via a secure system. In return, subscribers
receive economic benefit for their participation.
[0250] In another embodiment, the present invention is a profiling
product that operates at the point-of-purchase (retail outlet, mail
order, or other retail purchasing system) and produces profiles
based on the purchases of the subscriber. The specific purchases of
the consumer are not stored, and the profiles are only utilized by
authorized members.
[0251] In another embodiment, the principles of the present
invention are deployed as a secured credit card that may be
utilized to monitor purchase transactions of the subscribers and to
ensure consumers that their purchase information will not be
aggregated, but to allow them to gain the benefits of secure
profiling. By the use of a secured credit card, consumers may allow
profiling based on their purchase records. This embodiment ensures
that the raw transaction data (detailed purchase records) is not
stored.
[0252] By using a credit card that is part of the targeted
advertising business, it is possible to track the purchases made by
that consumer. Although it is preferable to discard the specific
transaction data after profiling, use of a credit card associated
with the targeted advertising process allows for tracking of
purchase activity by consumers who "opt-in". The credit card may
also be subsidized by the advertising dollars, thus creating a low
interest rate credit card, which would be an incentive to
"opt-in".
[0253] In this embodiment, advertisers may also be able to
correlate their advertisements against consumer information and
target advertisements to the subscribers, however, the advertisers
are not provided access to the profiles themselves. The revenues
generated by the credit card issuer/profiler may be used to
subsidize the credit card in the form of decreased interest rates
and/or discounts or rebates for use of the card. Another feature of
the secured credit card is the ability to determine if a displayed
advertisement resulted in the purchase of an item. As an example,
if a targeted advertisement is displayed to a consumer via the
present system and the item is subsequently purchased using the
secured credit card, the advertisement may be marked as effective.
On a statistical basis, the effectiveness of an advertising
campaign may be readily measured when the subscribers receive
advertisements through the secured system and make their purchase
using the secured credit card of the present invention.
[0254] In one implementation, the present invention may be based on
the use of a secure correlation server (SCS) connected directly to
an access platform, e.g., a Broadband Digital Terminal (BDT). In
this implementation, the secure correlation server is capable of
receiving video profiles (foiined from channel changes and dwell
times) from the BDT as well as receiving consumer purchase records
from participating retail outlets and/or online stores. The SCS may
also utilize the data from external databases.
[0255] In this implementation, the profiling is performed based on
consent, e.g., the profiles of subscribers/consumers who opt-in for
the service agree to have their demographic and preference profiles
stored on the SCS.
[0256] Advertisers wishing to send advertisements to a subscriber
during an ad opportunity (Web page ad location or video advertising
spot) transmit an ad characterization vector to the SCS. The ad
characterization vector may be created by the advertiser by simply
filling out a Web page containing questions (with pull down
answers) that describe the target market by demographic information
or by preference information.
[0257] Upon receiving the ad characterization vector the SCS
correlates the ad characterization vector with the
subscriber/consumer characterization vector. Based on the results
of this correlation, the SCS may determine whether the ad should be
delivered to the subscriber, or if an alternate ad should be
presented.
[0258] A privacy firewall may be maintained between the BDT and the
SCS to ensure that subscriber/consumer characterization vectors may
not be read or constructed by unauthorized parties. Because no raw
data (consumer purchase or viewing records) are stored on the SCS,
there is no possibility of unauthorized access of private
information. This system allows subscribers/consumers the ability
to receive more desirable advertisements while simultaneously
receiving discounts for Internet/video services and at
retail/online outlets. Advertisers receive the benefit of more
effective advertisements, and thus spend advertising dollars more
efficiently. This increase in efficiency results in increased
revenue stream. Advertisers pay a premium for targeted
advertisements, as opposed to traditional linked sponsorship
advertisements in which a flat rate is paid for access to an
audience whose characteristics are only generally known.
[0259] Although this invention has been illustrated by reference to
specific embodiments, it will be apparent to those skilled in the
art that various changes and modifications may be made, which
clearly fall within the scope of the invention. The invention is
intended to be protected broadly within the spirit and scope of the
appended claims.
* * * * *