U.S. patent application number 11/563329 was filed with the patent office on 2007-04-19 for target information generation and ad server.
This patent application is currently assigned to YODLEE.COM, INC.. Invention is credited to Subhash Sankuratripati, Dinesh K. Shanbhag, Jaideep Srivastava.
Application Number | 20070088821 11/563329 |
Document ID | / |
Family ID | 25248092 |
Filed Date | 2007-04-19 |
United States Patent
Application |
20070088821 |
Kind Code |
A1 |
Sankuratripati; Subhash ; et
al. |
April 19, 2007 |
Target Information Generation and Ad Server
Abstract
An advertisement selection and delivery system for selecting
advertisements based on profile information and rendering the
advertisements as accessible to a user operating a network-capable
appliance connected to a data-packet-network is provided. The
system comprises, a first server node connected to the network, the
first server node functioning as a user access point on the
network, a mass storage repository accessible to the first server
node, the repository for storing and serving user profile data, a
second server node connected to the network, the second server node
for generating user preference data, at least one advertisement
server connected to the network, the advertisement server for
serving advertisement data, a software application for generating
user preference lists and performing advertisement selection and at
least one network-capable appliance connected to the network the
network-capable appliance for receiving the advertisement data. A
user operating the network-capable appliance accesses the first
server node and receives the advertisement data, the advertisement
data selected for service by matching the user preference data to
stored advertisements and rendered accessible to the user during
the time of user access to the system from the network-capable
appliance.
Inventors: |
Sankuratripati; Subhash;
(Mountain View, CA) ; Srivastava; Jaideep;
(Sunnyvale, CA) ; Shanbhag; Dinesh K.; (Sunnyvale,
CA) |
Correspondence
Address: |
CENTRAL COAST PATENT AGENCY, INC
3 HANGAR WAY SUITE D
WATSONVILLE
CA
95076
US
|
Assignee: |
YODLEE.COM, INC.
Redwood Shores
CA
|
Family ID: |
25248092 |
Appl. No.: |
11/563329 |
Filed: |
November 27, 2006 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
09827011 |
Apr 4, 2001 |
7155508 |
|
|
11563329 |
Nov 27, 2006 |
|
|
|
09654320 |
Sep 1, 2000 |
|
|
|
09827011 |
Apr 4, 2001 |
|
|
|
Current U.S.
Class: |
709/224 |
Current CPC
Class: |
G06Q 30/02 20130101;
H04L 67/306 20130101; H04L 67/20 20130101; H04L 67/22 20130101 |
Class at
Publication: |
709/224 |
International
Class: |
G06F 15/173 20060101
G06F015/173 |
Claims
1. An advertisement selection and delivery system for selecting
advertisements based on profile information and rendering the
advertisements as accessible to a user operating a network-capable
appliance connected to a data-packet-network comprising: a first
server node connected to the network, the first server node
functioning as a user access point on the network; a mass storage
repository accessible to the first server node, the repository for
storing and serving user profile data; a second server node
connected to the network, the second server node for generating
user preference data; at least one advertisement server connected
to the network, the advertisement server for serving advertisement
data; a software application for generating user preference lists
and performing advertisement selection; and at least one
network-capable appliance connected to the network, the
network-capable appliance for receiving the advertisement data,
wherein a user operating the network-capable appliance accesses the
first server node and receives the advertisement data, the
advertisement data selected for service by matching the user
preference data to stored advertisements and rendered accessible to
the user during the time of user access to the system from the
network-capable appliance.
2. The advertisement selection and delivery system of claim 1,
wherein the data-packet-network is the Internet network.
3. The advertisement selection and delivery system of claim 2,
wherein the first server node and the second server node are in
addition to being connected to the Internet, connected to each
other by a separate dedicated network.
4. The advertisement selection and delivery system of claim 3,
wherein the software application is distributed in part on the
second server node and in part on the at least one advertisement
server.
5. The advertisement selection and delivery system of claim 4,
wherein the part of the software application executing on a second
server node directs generation of user preference lists and the
part of the software application executing on the at least one
advertisement server performs the advertisement selection according
to a user preference lists obtained from the second server.
6. The advertisement selection and delivery system of claim 3,
wherein the software application resides in whole and executes on
the second server node and advertisement selection is performed by
the second server node using advertisements delivered thereto from
the at least one advertisement server.
7. The advertisement selection and delivery system of claim 6,
wherein the second server node also serves the selected
advertisements.
8. The advertisement selection and delivery system of claim 2,
wherein the first server node is a cobranded server node servicing
clients of a cobrand partner to the entity hosting the system.
9. The advertisement selection and delivery system of claim 2,
wherein the advertisements include e-mail messages.
10. The advertisement selection and delivery system of claim 9,
wherein the advertisements include instant messages.
11. The advertisement selection and delivery system of claim 10,
wherein the advertisements include banner advertisements.
12. The advertisement selection and delivery system of claim 2,
wherein the network-capable appliance accesses the system through a
wireless networks.
13. The advertisement selection and delivery system of claim 2,
wherein the preference lists are generated using a knowledge base
data system.
14. The advertisement selection and delivery system of claim 13,
wherein the preference lists are used as search criteria in
conjunction with a search engine.
15. A preference-data generation server for generating preference
data using data mined from user profile data comprising: a data
port for receiving user profile data; a data port for accessing a
knowledge database; and a software application for mining, the user
profile data and for generating preference summaries by equating
the mined profile data to pre-established preference categories
listed in the knowledge database.
16. The preference-data generation server of claim 15, wherein the
preference summaries are generated in the form of categorized and
prioritized data lists.
17. The preference-data generation server of claim 16 further
comprising: a data port for receiving pre-configured advertisement
data; a data port for serving advertisement data; and a software
application for matching the advertisement data to individual ones
of generated data lists and for selecting the advertisement data
most closely matching the generated data lists for service.
18. The preference-data generation server of claim 17, wherein the
matching advertisement data is served to a network-access point
established on a data-packet-network.
19. The preference-data generation server of claim 18, wherein the
network-access point is a server node and the data-packet-network
is the Internet network.
20. The preference-data generation server of claim 16 further
comprising: a data port for serving the prioritized data lists.
21. The preference-data generation server of claim 20, wherein the
prioritized data lists are served to at least one advertisement
server operating on a data-packet-network.
22. The preference-data generation server of claim 21, wherein the
data-packet-network is the Internet network.
23. A method for dynamically serving advertisement data based on
user profile information to a user interface maintained on a
data-packet-network comprising the steps of: (a) compiling and
storing the user profile information on an ongoing basis; (b)
accessing the user profile information in order to mine the
information; (c) mining the accessed user profile information for
preference data; (d) formulating the preference data into a concise
summary-data list; (e) selecting pre-configured advertisements from
a database containing stored advertisements, the selection
accomplished by matching the advertisements to data contained in
the summary-data list; and (f) serving the selected advertisements
to the user interface.
24. The method of claim 23, wherein the data-packet-network is the
Internet network.
25. The method of claim 24 wherein in step (a), compilation of user
profile information is accomplished by recording user activity on
the network.
26. The method of claim 25 wherein in step (a), compilation of user
profile information is augmented through manual data procurement
methods.
27. The method of claim 25 wherein steps (b)-(f) are performed as a
sequence launched as a result of the profile the user connecting to
an accessing the user interface using a network-capable
appliance.
28. The method of claim 27 wherein in step (c), data mining is
accomplished through a parsing method.
29. The method of claim 28 wherein in step (c), the preference data
is categorized and prioritized according to pre-configured
preference categories related to types of advertisements.
30. The method of claim 29 wherein in step (d), the summary-data
list is of the form that can be propagated through the network.
31. The method of claim 30 wherein in step (d), the summary-data
list is sent to an advertisement server wherein the advertisement
server performs steps (e) and (f).
32. The method of claim 29 wherein steps (c)-(f) are accomplished
by a single server node connected to the network.
33. The method of claim 27 wherein in step (e), the advertisements
are generated to fit the summary-data list and are of the form of
instant messages advertisements.
34. The method of claim 27 wherein in step (e), the advertisements
are generated to fit the summary-data list and are in the form of
e-mailed advertisements.
Description
CROSS-REFERENCE TO RELATED DOCUMENTS
[0001] The present invention is a continuation-in-part (CIP) to
patent application Ser. No. 09/654,320 entitled "Method and
Apparatus for Multifaceted Profiling of Individual Cobrand Users"
filed on Sep. 1, 2000, disclosure of which is incorporated herein
in its entirety herein by reference.
FIELD OF THE INVENTION
[0002] The present invention is in the field of Internet-based
services and applications, and pertains more particularly to
methods for generating a concise data package containing target
information about a user for the purpose of directing ad generation
to an interface employed by the user.
BACKGROUND OF THE INVENTION
[0003] The information network known as the world-wide-web (WWW),
which is a subset of the well-known Internet, is arguably the most
complete source of publicly accessible information available.
Anyone with a suitable Internet appliance such as a personal
computer with a standard Internet connection may access (go
on-line) and navigate to information pages (termed web pages)
stored on Internet-connected servers for the purpose of gathering
information and initiating transactions with hosts of such servers
and pages.
[0004] Often times, in order to improve the quality of services
offered by a particular website, it is important to understand user
activity in relationship to that site. This is to say that while a
user is navigating through a website, obtaining a dynamic profile
of the user's habits, activities and personal information would
prove beneficial to the overall improvement of a service providing
or commercial website. In addition to utilizing user profiles for
website service-enhancement purposes, companies routinely pay for
such information in order to better target users for advertising
and marketing purposes.
[0005] In a cobrand relationship known to the inventor, cobrand
partners contract with a service-providing entity in order to
provide Internet services offered by the entity. The cobranded
services are made available to subscribers of the cobrand partners
through dedicated servers maintained by the service-providing
entity. Users who subscribe to such services typically have at
least some personal profile information known to the cobrand
partners through their normal subscription and interaction
activities. In addition, a service-providing entity may track
certain information about users who are accessing and interacting
with cobranded services maintained by the service-providing entity.
For example, information such as types of products purchased, types
of web pages accessed at service sites, frequency of buying, time
spent at sites, and so on, may be tracked and stored in a secure
database by the service-providing entity. This is made possible by
the fact that the service-providing entity maintains and provides
the services and the equipment through which the services are made
available.
[0006] There are a variety of known methods for obtaining
information about individual users who visit websites online. Some
commonly known methods are sending and retrieving interactive
cookies, conducting on-line surveys, parsing completed online
forms, recording purchase histories, and many other techniques. A
typical user profile automatically compiled by a Web company is
limited to information that can be obtained from the user while at
one of the company-sponsored sites, or through interacting with the
user during registration processes. As such, the profile is not
complete or well rounded and tends to reflect content related to
the nature of business conducted by the Web company. For example, a
purchase history compiled by a Web-based clothing retailer is
limited to the subject of clothing. In order to obtain a
well-rounded profile of an individual that covers a variety of
topics, information must be bought, sold, or traded between Web
companies doing business on the Internet. It is known in the art
that there are many companies in existence that specialize in
information brokering. In the case of cobranding, where the
service-providing entity provides proxy navigation and data summary
services for users, data about a user's activity related to
interaction with cobrand services includes data related to a
plurality of disparate Web-sites, which are involved in some aspect
of the cobrand services. It has occurred to the inventor that much
information may be automatically obtained about users from user
interaction and proxy interaction with many Web sites without being
required to obtain the data through purchase or trade with
companies hosting Web-sites involved in cobranded services.
[0007] A system known to the inventor and taught in the related
document listed in the cross-reference enables automated collection
of data about a user through monitoring user interaction on the
network. The data-collection system includes a proxy server
connected to the data-packet-network for providing proxy services
and for monitoring user access and interaction with those services,
a dedicated server interface connected to the network for providing
user access to the proxy services, and a software application
ruling on the proxy server for collecting and storing data obtained
as a result of active user-interaction with the proxy services. In
preferred embodiments of the invention, the data is collected in an
automated fashion and is used for the construction of multifaceted
user profiles, which may be periodically updated in an automated
fashion as a result of continued user interaction with provided
proxy services through the dedicated server interface.
[0008] The system also incorporates manual techniques used in data
collection and integrates results obtained manually with those
obtained in an automated fashion to compile detailed profiles of
individual users. One of the uses of a complete and detailed
profile on a user is to incorporate the compiled information for
use in advertising as is generally known in the art. However, in an
automated network environment ads must be delivered into
user-operated interfaces as accessed web pages are loaded. In
current art, ads delivered according to profiling, either text or
graphic, are more or less static in that they do not change in
content to the extent that a user may change in personal habits,
preferences, or other profiled attributes. These ads are decided on
based on an overall picture of a user or a group of users.
Therefore, they are not really as flexible or target-oriented as
might be desired by both users and advertising companies.
[0009] It has occurred to the inventors that through further
innovation and refinement, an automated profiling system may be
adapted to define a user's profile in such a way as to incorporate
slight changes in content, categories, and preferences as they are
discovered. However, in order to cause ads to be delivered such
that they incorporate evolving changes, a system must be developed
to communicate the mean of those changes in a way that may be
utilized on the fly as Web pages delivered into a user interface
are loaded.
[0010] What is clearly needed is a system for packaging and
communicating profile data to ad sources such that dynamic ads may
be selected and delivered based on mean changes in the profile
data. Such a system would provide a much greater degree of
compliance of delivered ads to a user's preferences and status
states enabling a greater hit rate and a greater profit margin for
ad companies.
SUMMARY OF THE INVENTION
[0011] In a preferred embodiment of the present invention, an
advertisement selection and delivery system for selecting
advertisements based on profile information and rendering the
advertisements as accessible to a user operating a network-capable
appliance connected to a data-packet-network is provided. The
system comprises, a first server node connected to the network, the
first server node functioning as a user access point on the
network, a mass storage repository accessible to the first server
node, the repository for storing and serving user profile data, a
second server node connected to the network, the second server node
for generating user preference data, at least one advertisement
server connected to the network, the advertisement server for
serving advertisement data, a software application for generating
user preference lists and performing advertisement selection, and
at least one network-capable appliance connected to the network the
network-capable appliance for receiving the advertisement data.
[0012] In a preferred use, a user operating the network-capable
appliance accesses the first server node and receives the
advertisement data, the advertisement data selected for service by
matching the user preference data to stored advertisements, which
are rendered accessible to the user during the time of user access
to the system from the network-capable appliance.
[0013] In a preferred embodiment, the system is implemented on the
Internet network. In addition to being connected to the Internet,
the first server node and the second server node are, in one
embodiment, connected to each other by a separate dedicated
network. The software application is, in one implementation,
distributed in part on the second server node and in part on the at
least one advertisement server. In this aspect, the part of the
software application executing on a second server node directs
generation of user preference lists and the part of the software
application executing on the at least one advertisement server
performs the advertisement selection according to a user preference
list obtained from the second server.
[0014] In another aspect, the software application resides in whole
and executes on the second server node and advertisement selection
is performed by the second server node using advertisements
delivered thereto from the at least one advertisement server. In
this aspect, the second server node also serves the selected
advertisements, functioning as an ad broker. Also in some aspects,
the first server node is a cobranded server node servicing clients
of a cobrand partner of the entity hosting the system. In this
aspect, the advertisements may include e-mail messages or instant
messages. However, in preferred aspects, the advertisements served
are banner and text advertisements.
[0015] In one aspect, the network-capable appliance accesses the
system through a wireless network. In another aspect of the system,
the preference lists are generated using a knowledge base data
system. In still another aspect, the preference lists are used as
search criteria in conjunction with a search engine.
[0016] In another aspect of the present invention, a
preference-data generation server for generating preference data
using data mined from user profile data is provided. The server
comprises, a data port for receiving user profile data, a data port
for accessing a knowledge database and a software application for
mining the user profile data and for generating preference
summaries by equating the mined profile data to pre-established
preference categories listed in the knowledge database. In all
aspects, the preference summaries are generated in the form of
categorized and prioritized data lists.
[0017] In one embodiment, the preference-data generation server
further comprises, a data port for receiving pre-configured
advertisement data, a data port for serving advertisement data and
a software application for matching the advertisement data to
individual ones of generated data lists and for selecting the
advertisement data most closely matching the generated data lists
for service. In a preferred aspect, the matching advertisement data
is served to a network-access point established on a
data-packet-network, which, in preferred instances, is the Internet
network. The preference-data generation server, in one aspect,
further comprises a data port for serving the prioritized data
lists. In this aspect, the prioritized data lists are served to at
least one advertisement server operating on a data-packet-network,
which is the Internet network. Also in this aspect, the process of
selecting advertisements is performed by the at least one
advertisement server.
[0018] In another aspect of the present invention, a method for
dynamically serving advertisement data based on user profile
information to a user interface maintained on a data-packet-network
is provided. The method comprises the steps of, (a) compiling and
storing the user profile information on an ongoing basis, (b)
accessing the user profile information in order to mine the
information, (c) mining the accessed user profile information for
preference data, (d) formulating the preference data into a concise
summary-data list, (e) selecting pre-configured advertisements from
a database containing stored advertisements, the selection
accomplished by matching the advertisements to data contained in
the summary-data list and (f) serving the selected advertisements
to the user interface.
[0019] In a preferred embodiment, the method is practiced on the
Internet network. In a preferred aspect of the method in step (a),
compilation of user profile information is accomplished by
recording user activity on the network. In another aspect, in step
(a), compilation of user profile information is augmented through
manual data procurement methods. In a preferred aspect of the
method, steps (b)-(f) are performed as a sequence launched as a
result of the profiled user connecting to and accessing the user
interface using a network-capable appliance.
[0020] Further to the above, in step (c), data mining is
accomplished through a parsing method. Also, in step (c), the
preference data is, in preferred embodiments, categorized and
prioritized according to pre-configured preference categories
related to types of advertisements. In one aspect of the method, in
step (d), the summary-data list is of the form that can be
propagated through the network. In this aspect, in step (d), the
summary-data list is sent to an advertisement server wherein the
advertisement server performs steps (e) and (f). In still another
aspect, steps (c)-(f) are accomplished by a single server node
connected to the network. In one aspect of the method in step (e),
the advertisements are generated to fit the summary-data list and
are of the form of instant messages advertisements. In another
aspect, in step (e), the advertisements are generated to fit the
summary-data list and are in the form of emailed
advertisements.
[0021] Now, for the first time, a system for packaging and
communicating profile data to ad sources such that dynamic ads may
be selected and delivered based on mean changes in the profile data
is provided. Such a system provides a much greater degree of
compliance of delivered ads to a user's preferences and status
states enabling a greater hit rate and a greater profit margin for
ad companies.
BRIEF DESCRIPTION OF THE DRAWING FIGURES
[0022] FIG. 1 is an overview of a communication network wherein
cobrand user-profiling is practiced according to an embodiment of
the present invention.
[0023] FIG. 2 is a block-diagram illustrating various data
categories and data-gathering methods used to create a multifaceted
user-profile according to an embodiment of the present
invention.
[0024] FIG. 3 is an overview of a communication network wherein
target advertisement based on evolutionary profiling is practiced
according to an embodiment of the present invention.
[0025] FIG. 4 is a process-flow diagram illustrating various steps
for profiling and ad serving based on received profile information
according to an embodiment of the present invention.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0026] FIG. 1 is an overview of a communication network 9 wherein
cobrand-user profiling is practiced according to an embodiment of
the present invention. Communication network 9 contains all of the
equipment and network connections required in order to establish a
functional network-communication capability.
[0027] In a preferred embodiment of the present invention,
communication network 9 is implemented on the Internet network,
which is represented herein by an Internet backbone 11. Internet
backbone 11 represents all of the lines, connection points, and
equipment that make up the Internet network as a whole. Therefore,
there is no geographic limit to the practice of the present
invention.
[0028] A plurality of cobrand servers (CBS) 23, 25, and 27 are
illustrated, in this example, as connected to Internet backbone 11.
CBS 23-27 are adapted as Internet file servers dedicated to users
of cobrand services provided by a service-providing company, which
also maintains the servers. A main server 15, also illustrated as
connected to Internet backbone 11, is adapted as a data-aggregation
and distribution source and is hosted by the same service provider
hosting servers 23-27. Cobrand servers 23-27 are dedicated to
cobrand partners and their subscribers whom have elected to access
services offered by the service provider, which hosts server
15.
[0029] In addition to data-aggregation and distribution services
(cobrand service) server 15 also functions, in this example, as an
activity-tracking server tracking on-line activity of cobrand
users. Companies providing cobrand services to their
subscriber-ship typically specialize in providing search
capabilities and limited portal services. The companies hosting CBS
23-27 may be termed cobrand partners of the described
service-providing company. The nature of services provided by CBS
23-27 will depend on the nature of agreements forged between
various cobrand partners and the service-providing entity. For
example, some of CBS 23-27 may allow users to obtain e-mail from
various e-mail servers. Likewise, each of CBS 23-27 may provide
varying functionalities generally related to data collection,
aggregation and summary services. It is sufficient to say that
subscribers to cobranded services interface with CBS 23-27 in order
to receive such services.
[0030] A plurality of content servers (CS) 19 and 21 are adapted as
Internet data servers hosted by companies contracted to provide
specific content to the service-providing entity, which makes the
content available through CBS 23-27. Servers 19 and 21 are
illustrated herein as connected to Internet backbone 11.
[0031] CS 19 and 21 are dedicated to providing specific Web content
such as weather information, stock quotes, financial news,
entertainment news, and so on. There are many possibilities as to
the nature of the content provided by servers 19 and 21. In some
cases content provided by servers 19 and 21 may be inaccessible
without subscription or membership. It is noted herein that CS 19
and 21 are not, in this example, hosted by the entity hosting main
server 15 and cobrand servers 23-27 rather, they are hosted by
companies contracting with the service-providing entity of this
example and provide content according to contract stipulation.
[0032] A plurality of Free (FS) content servers 29 and 31 are
illustrated herein as connected to Internet backbone 11. FS 29 and
31, located to the right of CBS 23-27, in this example, and are
also adapted as Internet data servers, but in this case are not
providing content to CBS 23-27. FS 29-31 are not in any way
associated with the entity hosting main server 15, however, in some
cases may be accessed through main server 15 by proxy such that
accessed content may also be tracked by main server 15. Like CS
19-21, there are many possibilities as to the nature of content
provided by FS 29-31, the term "free" is used in this example to
convey that the services and/or content provided within FS 29-31 is
available to anyone who accesses it.
[0033] It will be apparent to one with skill in the art that there
may be many more CBS, CS, and FS illustrated in this example
without departing from the spirit and scope of the present
invention. The inventor illustrates only a few of each class of
server in this example and deems the illustration sufficient for
explanation of the present invention. It is repeated here that CBS
23-27 are cobrand servers maintained by a service-providing entity
also hosting main server 15. CS 19-21 are content servers hosted by
companies contracting with the described service-providing entity
to make their content available to CBS 23-27. FS 29-31 are free
servers not affiliated in any way with the service-providing
entity.
[0034] A plurality of cobrand subscribers/users 17 (within dotted
rectangle) is illustrated as having Internet connection to Internet
backbone 11. Internet connection in this example includes all of
the known methods for accessing the Internet network. An
internet-service-provider (ISP) is not detailed in this example,
but may be assumed to be present in a scenario where users 17 are
accessing Internet 11 through normal dial-up modem technology,
which is most common. Other methods include wireless modem
connection, cable modem connection, and so on.
[0035] In this example, users 17 are illustrated as operating
personal computers (PC) to access Internet 11. In actual practice,
any Internet-capable appliance may be used to practice the present
invention as long as it has network-browsing and display
capabilities. Each user 17 may freely navigate to and interact with
CBS 23-27, FS 29-30, or CS 19-21. However, when users are accessing
cobrand services from one of CBS 23-27, at least part of the
service enables them to have data obtained from any of CS 19-21 or
FS 29-31. For example, if a user 17 is logged into CBS 27 to
receive cobrand services, specific data requested by the user such
as weather or financial news would be obtained by proxy from CS
19-21 because of contract stipulation. If data is requested from
any of FS 29-31, then the requesting user must provide information
such as a URL address and type of data required to enable proxy
data collection and presentation because there is no affiliation
between FS 29-31 and CBS 27.
[0036] Main server 15, as previously described, can log the
activities of each of users 17 when they are interacting with CBS
23-27. Furthermore, server 15 has the ability to record activity
information related to ally proxy data request involving any other
servers navigated to on behalf of users 17. As a result, much data
specific to a user's activity may be obtained without giving notice
to or requesting data from companies hosting servers 19-21 or
servers 29-31.
[0037] An instance of software (SW) is provided to execute on main
server 15. SW at server 15 is adapted to record any user-activity
data routed through server 15. Therefore, data obtained through
cobrand services offered by CB 23-27 may be logged and identified
to particular users, and mined for data to include as profile data.
Server 15 may record types of content requested, description and
class of items purchased, nature and description of Web-sites
targeted for data requests, frequency of same type requests, lists
containing URLs and descriptions of user-registered Web sites, and
so on. Data about a cobrand user's on-line activity and behavior is
compiled and organized within server 15 and then stored as a part
of that user's multifaceted profile.
[0038] Other information about users may be obtained from companies
hosting CBS 23-27. Such information may include personal
information related to subscription and registration to receive
cobrand services, information obtained through registration and
interaction with a cobrand partners regular services routinely
accessed by cobrand users, and so on. In this way, much of a
multifaceted profile can be automatically generated and stored for
cobrand users 17. If a user is a frequent cobrand user and is
particularly prolific with on-line activity, then many facets about
that user's activity and behavior may be learned and recorded.
Profiling a user in this manner greatly reduces the need for
trading or purchasing partial profiles compiled by a plurality of
un-related sources. However, a multifaceted profile may, of course
be enhanced by supplementing the profile with purchased or traded
data originating from out-side sources.
[0039] Significant market advantages exist for an entity that can
obtain a multifaceted profile on each of its clients. For example,
many user profiles may be mined for more specific data, which may
be generated into list-reports detailing preferences and activity
traits of many users. Such lists may be created and tailored for
advertisement companies or other service providers willing to pay
for the information. Thresholds and special rules may also be
implemented during profile configuration and maintenance such that
profile information may be automatically updated over prolonged
user patronage of cobrand services.
[0040] FIG. 2 is a block-diagram illustrating various data
categories and data gathering methods used to create a multifaceted
user-profile according to an embodiment of the present invention.
In this example, there are three basic categories of data used to
create a dynamic multifaceted profile. These categories are
illustrated in this example, by element number 37 (demographic
info), element number 39 (account info), and element number 41
(on-line behavior). Element number 43 represents various
data-gathering methods, which may be used to obtain data described
in the data categories mentioned above.
[0041] Every cobrand user is supplied with a cobrand identification
(ID) and a member ID. In this way, data obtained and warehoused
from internal or external sources is easily identifiable to a
particular user. Profiles may be automatically assembled using this
warehoused data according to enterprise rules. It is important to
note herein that each data category 37, 39, and 41 may be populated
using automatic data-gathering methods (element 43) such as data
capture during interactive sessions. Data capture, as is used in
this specification, means recording any and all data about a user
during an on-line session, which includes any proxy services. Of
course, some data may be supplied by purchasing from the outside,
trading, or sharing with partners. These additional options are
included under data-gathering methods 43. For the most part
however, automated data capture should be sufficient for supplying
a viable multifaceted profile in most cases. This is especially
true if a user is prolific in his or her patronage of cobranded
services.
[0042] Referring now to demographic info 37, there is illustrated a
plurality of subcategories, which will be discussed from top to
bottom. Income level is one important subcategory of demographic
information. By knowing a user's income level, advertisements for
travel, financial services, and consumables may be more properly
tailored for the individual. Location information may also be used
to enhance local advertising.
[0043] Family relationships are important for understanding
lifestyle characteristics, identifying future consumers, and so on.
Gender, age, and income levels of family members may also be
important in creating a multifaceted profile. Hobbies and other
preferences may also be included in data gathered for demographic
information. Such data also contributes to understanding lifestyle
characteristics and identifying products and services that may fit
a user.
[0044] Demographic information may be obtained through recording
on-line purchase events, registration events, and from general
population of on-line forms. Therefore, most demographic
information may be obtained through automated data-capturing
techniques. In some cases, companies contracting for cobrand
service-enhancement with a service-providing entity may simply
forward or share some demographic information. Such an arrangement
may be, in some embodiments, required as part of contract
negotiation. In other cases, especially if certain users are not
prolific in on-line activity, demographic information may be
purchased from the outside or obtained through trade with outside
organizations, Therefore, information that cannot be obtained
through data capture may be obtained through other methods in order
to supplement any relatively weak profiles.
[0045] Account information 39 is illustrated herein as divided into
4 basic subcategories. These subcategories are listed from top to
bottom as entertainment, business, financial, and investment. Each
category may be further divided into more subcategories has deemed
appropriate. Account information represents data obtained from
on-line accounts belonging to a particular user. A user may
subscribe to many of these accounts and may add them to cobrand
service sites for the purpose of being able to access information
from such accounts without physically navigating to them. As proxy
services are performed on behalf of a user concerning a user's
registered accounts, data used in accessing the accounts and data
returned as a result of task performance is collected and
incorporated as profile data.
[0046] As data is automatically compiled about a user over time,
the user's profile becomes more and more valuable and accurate.
After a period of time, the service-providing entity maintaining
the cobrand services and the user's profile may generate automated
reports detailing certain aspects of the user's profile for
selective distribution to paying clients. Secure information such
as credit card numbers, Social Security numbers, personal
identification numbers, passwords, and the like remain in a state
of data encryption, or otherwise deleted from data reports
containing profile information.
[0047] On-line behavior is compiled using user-activity and
server-activity data. Such raw data is collected and analyzed in
order to compile an on-line behavior profile. Subcategories of user
activity that may constitute on-line behavior illustrated within
block 41 are listed and discussed from top to bottom.
[0048] Identification of Web sites visited either directly or
through proxy services maybe automatically captured. Types of
products purchased from those sites may also be automatically
captured as well as frequency parameters associated with purchases
illustrated herein as a subcategory of products purchased.
Frequency of purchases may simply mean how often a purchase is made
over a number of visited sites. The frequency of purchases may also
mean the frequency of purchase of one particular product.
[0049] Time accessing cobrand services may also be included and
incorporated in forming an on-line behavior profile. Parameters
surrounding banner-ad or sponsor clicking may be incorporated in
order to determine certain preferences. Finally, on-line histories
may be created and maintained on virtually any category or
subcategory associated with blocks 39 and 41.
[0050] In a preferred embodiment of the present invention, most if
not all of the data compiled about a user is collected using
automated data capturing techniques implemented during the normal
course of the user accessing cobrand services. The very nature of
such services enables many of these data capturing techniques to be
utilized.
[0051] It will be apparent to one with skill in the art, that there
may be more categories and subcategories included in this example
without departing from the spirit and scope of the present
invention. The inventor has outlined basic categories and basic
subcategories and deems them sufficient for illustrative purposes.
Therefore, the inclusion of such subcategories and categories in
this example should not be construed as a limitation in any
way.
[0052] It will also be apparent to one with skill in the art, that
the unique cobrand architecture implemented between the
service-providing entity of FIG. 1 represented by a server 15 and
the plurality of cobrand partners utilizing CBS 23-27, which are
maintained by the same entity, enables multifaceted profiling of
users to be accomplished in automated fashion. Therefore, much
manual labor and research is eliminated from the data profiling
process.
Target Information Generation and Ad-Serving
[0053] FIG. 3 is an overview of a communication network 45 wherein
target advertisement based on evolutionary profiling is practiced
according to an embodiment of the present invention. Communication
network 45 includes a sub-network 49, which in this example is the
well-known Internet network and will hereinafter be referred to as
Internet 45. Internet 45 is the preferred medium used to
interconnect various other components, which cooperate and
communicate with each other through Internet 45 for the purpose of
accomplishing the goal of the present invention, which is to
provide a high level of ad compliance to user profiled states.
[0054] In a preferred embodiment of the present invention, Internet
49, represented herein as a cloud, has a service provider 50
illustrated therein and adapted to provide a unique service that
benefits users of the service, which includes users and
advertisers. Service provider 50 comprises a file server 51
illustrated as connected to an Internet backbone 47. Internet
backbone 47 represents all lines, connection points and equipment
making up the Internet network as a whole. Therefore, there is no
geographic limitation the practice of the present invention. File
server 51 is adapted as a portal server is so labeled. Server 51
represents the main user interface of service provider 50. A mass
storage repository 53 is illustrated within the domain of service
provider 50 and connected to server 51 by virtue of a direct data
link. Repository 53 is adapted to hold user profile information as
so labeled. User profiles stored in repository 53 are generated in
part at server 51 by recording user activity at the server.
[0055] A file server 55 is illustrated within the domain of service
provider 50 and is connected to backbone 47. Server 55 is adapted
in part as a target ad server as so labeled, and in part as an in
information generation server. A mass storage repository 57 is
illustrated within the domain of service provider 50 and connected
to server 55 by a high-speed data link. Repository 57 is adapted to
store preference categories as so labeled. The term preference
categories is defined in this specification as categories of user
preferences mined from the user profiles held in repository 53.
[0056] Servers 51 and 55 are interconnected, in this example, by a
separate high-speed data network 52. However, such connection is
not required in order to practice the present invention as servers
51 and 55 may communicate with each other over Internet backbone
47. Mass repositories 53 and 57 are illustrated in this example as
external repositories. However, in other embodiments, they may be
internal repositories within their respective servers. Also in
other embodiments, one repository may be used as long as it is
accessible to both servers 51 and 55.
[0057] A plurality of ad partners 62 is illustrated within the
domain of Internet network 49 and adapted via software and
equipment to serve advertisements to target interfaces over
Internet backbone 47 and through any sub-networks represented by
backbone 47. Ad partners 62 cooperate with service provider 50 in
order to provide dynamic target advertising to both wireless device
users illustrated herein by a dotted rectangle 65, and connected PC
users represented herein by a dotted rectangle labeled 63.
[0058] In this example, there are two ad servers illustrated within
the domain of ad partners 62. These are server 59 and server 61.
Servers 59 and 61 are both connected to Internet backbone 47. It is
important to note herein that servers and 59 and 61 are adapted to
serve advertisements and may be hosted by separate ad companies
making up ad partners 62. It will be apparent to one with skill in
the art that there may be many more ad servers represented in this
example without departing from the spirit and scope of the present
invention. The inventor illustrates only two ad servers in this
example and deems the illustration sufficient for explanation of
the present invention.
[0059] Wireless device users 65 represents users who connect to
Internet backbone 47 through Internet-capable appliances such as
handheld computers and cellular telephones represented herein as
icons illustrated within the domain of users 65. PC users 63
represents users who connect to backbone 47 via a wired connection
method such as the well-known dial-up/modem method. Sub-networks
such as the well-known public-switched-telephony-network (PSTN) and
cellular networks are not illustrated in this example but may be
assumed to be present in represented in general by backbone 47,
which includes connected sub-networks. Similarly,
Internet-service-providers (ISPs) and network gateways adapted to
bridge wireless networks with Internet 49 are not illustrated but
may be assumed to be present.
[0060] In this example, it is assumed that users 65 and users 63
login to Internet backbone 47 and access portal server 51 to
receive various data-aggregation and summaries services known to
the inventor. As such, user profiles represented in mass repository
53 are profiles compiled from demographic information and activity
information obtained from users 65 and 63 on an ongoing basis. Such
user profiles may be assumed to be the multifaceted profiles
described with reference to U.S. patent application Ser. No.
09/654,320 listed in the cross-reference section of this
specification. A goal of the present invention is to summarize and
package user profiles contained in repository 53 into manageable
data lists containing preference categories illustrated in
repository 57. This is accomplished by a software (SW) application
56 provided to execute, in this example, on server 55.
[0061] Multifaceted profiling of users may include many user
aspects including but not limited to data related to queries
initiated; types of data collected from various data requests; the
nature and description of Web-sites accessed; and access of
frequently asked questions (FAQs) pertaining to a user's particular
interests. Information concerning the data and/or product
providers' response, behavior, on-line senescence, and other
characteristic data of providers and services is learned and
recorded by the profiling software. A user's multifaceted profile
constantly evolves by virtue of learning new information about a
user as profiling software continually tracks and updates recording
new facets of user activity and online behavior.
[0062] SW 56 is adapted to access the multifaceted user profiles
contained in mass repository 53 by virtue of high-speed data
network 52 and is adapted to generate concise data lists of
categorized, and in some cases, prioritized user-preference
categories. The preference lists are generated to be small enough
to easily be propagated through Internet backbone 47. In one
embodiment, SW 56 uses knowledge-based and data parsing techniques
in order to summarize a multifaceted profile into a categorized and
prioritized list of preferences for advertising purposes.
[0063] A software (SW) application 60 is provided to execute on ad
servers hosted by entities making up ad partners 62. In this
example, an instance of SW 60 resides on server 59 and an instance
of SW 60 resides on server 61. SW 60 is, in a preferred embodiment,
adapted to search for and deliver ads based on summary lists of
categorized preferences delivered to respective host file servers
59 and 61 from server 55, which in this example, functions as an ad
broker. Ads delivered to server 55 from servers 59 and 61 are
selected according to the appropriate theme or themes defined in
the categorized preference lists used to invoke the ad
delivery.
[0064] The ads received at file server 55 are rerouted through
portal server 51 into appropriate Web interfaces accessed there
from by users 65 and 63 using their respective appliances. This
unique ad service is dynamic in nature meaning that a new,
summarized preference list is generated each time one of users 65
or 63 logs into a personalized Web page served by server 51. The
generated preference list is, of course, derived from that user's
multi-faceted profile held in repository 53. During the time
required to download a personalized page from server 51, the
preference list is propagated to one or both of servers 59 and 61
where it is used as the criteria for selecting appropriate ads for
routing into the personalized page. As a user's multifaceted
profile evolves, so to does an associated preference list.
[0065] In another embodiment, server 55 may be adapted to function
as a target information generation server, but not as an ad broker.
In this case, ads delivered from server 59 and 61 would be
propagated directly into server 51 during the appropriate time of
Web-page service to a requesting user. However, in this embodiment
preference lists are generated and sent to ad servers 59 and 61 at
the beginning of page access by users.
[0066] Internet 49 is, in this example, the preferred data medium
that hosts the data-aggregation and distribution system that is
used to convey the preferred ad data to users via cable/modem,
dial-up/modem, or satellite/modem transmission capabilities. In the
embodiment depicted herein, the distribution of ad data may be made
to any of users 65 employing wireless appliances such as the ones
illustrated within the domain of box 65, or to new devices being
made available or in development at the time of this writing.
Likewise, ad distribution may be accomplished via Internet 47 or
the Internet and cable/modem networks to users of personal
computers, Web TVs, computerized workstations, and other computer
embodiments having network browsing and display capabilities. All
that is required to practice the invention is a user have an
appliance capable of accessing backbone 47 and server 51 through
any one of a variety of known Internet-access methods,
[0067] In one embodiment of the present invention, users 65 and 63
may access the unique, dynamic ad service by connecting to any one
of a plurality of cobranded-interface servers represented in FIG. 1
of Ser. No. 09/654,320 as cobrand servers (CBS) 23-27. It is noted
herein, that cobranded servers may be utilized in conjunction with
a main portal server (51) as user interfaces to which ads may be
delivered.
[0068] In one embodiment of the present invention SW 56 executing
on server 55 within the domain of service provider 50 has the
capability of generating categorized preference lists and the
capability of polling ad sources (servers 59 and 61) for
appropriate ads. This may be accomplished through an established
prioritization scheme developed by service provider 50 and ad
partners 62. In this case, ads delivered into a particular
interface may be sourced from more than one server. The exact
prioritization scheme will depend on the nature of agreements
forged. In this embodiment, preference categories held in
repository 57 may not actually be sent to servers 59 or 61 of ad
partners 62. Rather, they may be used as search criteria for a
search engine implemented as part of SW 56. Such a search engine or
polling software may be adapted to identify and retrieve ads from a
plurality of ad servers represented by server 59 and 61 with the
ads rated by percentage of match to the preference criteria.
[0069] According to another embodiment of present invention SW 60
may be enhanced with a search engine adapted to search ads hosted
by a host server 59 or 61. In this case, a categorized preference
list is sent to an appropriate ads server (59, 61) according to
contractual arrangement where in an internal search comprising ad
identification and delivery is performed based on the received
preference criteria.
[0070] A software application (not shown) provided to execute on
server 51 operates in conjunction with mass storage repository 53
to obtain multi-faceted user profiles held in repository 53 (U.S.
Ser. No. 09/654,320). As described in an embodiment of related
application U.S. Ser. No. 09/654,320 such an application has the
ability to record activity information related to any data request
involving any of the servers navigating ad partners, data sources,
and message traffic. In addition, the capability includes the
ability to aggregate and categorize types of data requested and/or
collected, types of inquiries initiated and data provided, and the
variable categories and cost envelopes of purchases and/or
deliveries. Moreover, identification and recording of the different
types of interface devices employed by users 63 and 65 is also
included in multi-faceted profiling.
[0071] It will be apparent to one with still in the art that SW
applications 56 and 60 may be provided of varying capabilities
without departing from the spirit and scope of the present
invention. For example, whether SW 56 includes polling
capabilities, or whether preference lists are simply delivered to
ad sources whereupon an internal ad search and delivery is
conducted is left to practical design and implementation. The
capability of delivering dynamic ads according to evolving user
preferences wherein such ads are decided on during user-login and
download of a WEB page represents a contribution to the art that
is, at the time of this writing, not available.
[0072] FIG. 4 is a process-flow diagram illustrating various steps
for dynamic ad serving based on received profile information
according to an embodiment of the present invention. At step 67,
multi-faceted profiling is collected and recorded at the location
of the service provider (50). Step 67 is an ongoing process
including collection and recording of user data, on-line behavior
characteristics, Internet activity, and preferences. The initial
data is obtained from user-supplied information and stored by the
service provider in mass storage repository 53 connected to portal
server 51 of FIG. 3.
[0073] Initial profile data is augmented each time a user initiates
on-line activity by supplementing the basic profile data with
information related to user generated queries, investigations,
purchases, on-line recreational activities and so on. It is noted
herein that in addition to profile data obtained by virtue of
automated process, profile data obtained from manual process may
also be included as referenced in U.S. patent application Ser. No.
09/654,320.
[0074] At step 69, a specific user profile is accessed and inclined
for summary data, which is organized into a categorized and, in
some cases, prioritized preference list. Step 69 results from a
user request and process of logging into a personalized Web
interface served by portal server 51 of FIG. 3, or from a described
cobrand server functioning as the interface. Data (profile) access,
mining and organization is accomplished by software (SW 56) through
the use of a knowledge-base technique or other rules-base scheme.
For example, a knowledge database may contain a plurality of listed
categories and descriptive keywords associated with those
categories. Category titles may be adopted into a preference list
by priority of number of matching hits to descriptive words
contained in a multifaceted profile.
[0075] The results obtained in step 69 will be slightly different
each time the process occurs by virtue of continual updating and
purging of a user's multifaceted profile. Incorporation of new user
preferences, traits, habitual or spontaneous site visits, ad or
sponsor clicking and so on this ultimately reflected in a
categorized preference list generated in step 69.
[0076] At step 73, a categorized preference list is temporarily
stored in a repository (repository 57, FIG. 3) connected to a
target ad server (server 55, FIG. 3). It is noted herein, that step
73 reflects one embodiment wherein preference lists are not sent
out to ad servers but are held at the service provider. Also at
step 73, ads received from various ad partners in step 81 are
collected and classified into appropriate preference categories and
stored in an ad database, which may be searched using a preference
list as a search criteria and a search function provided as part of
SW 56. In this case, ad identification and prioritized association
with a received preference list is performed by one machine within
the domain of the service provider.
[0077] According to another embodiment, step 73 comprises
intermediate receipt and propagation of preference lists to ad
servers (59 and 61 of FIG. 3) hosted by ad partners (62 of FIG. 3)
wherein ad identification, association and delivery according to
data contained in a preference list is performed by distributed
software (60 of FIG. 3). The preference category and user profile
information 71 are continually updated and provided to the target
ad server where the user profiles are catalogued and matched with
the products and services of the ad partners, FIG. 3 element
59.
[0078] Referring now back to the process described in this example,
ads generated by ad partners in step 81 are made available to
target ad server (51) by various agreements forged between the
parties involved. The provided ads can be associated under generic
categories or can be, in some cases, personalized to various users
or groups of users based on demographic data obtained previously
obtained during repeated execution of dynamic ad serving. Available
ads are limited only by the agreements between the ad partners and
the service providers and in all cases are established with the
needs and preferences of the users in mind. The products and
services of the ad partners are made available as pop-up banner
ads, static banner ads, pop-up text ads, static text ads, or other
inventive techniques developed by the parties involved.
[0079] Step 73 provides, in this example, both the receipt of
dynamic user-preference lists and the classification and collation
of ads into preference categories (ongoing) and matching of
appropriate ads to preference lists in process. Matching of stored
ads to received preference lists is accomplished using a database
search function. In one embodiment, the search function employs a
priority-coding scheme to expedite matching of ads to
user-preference lists.
[0080] At step 75, the target ad server serves selected ads to the
appropriate user interface identified in the particular
user-preference list. In this example, the target ad server (server
51 of FIG. 3) functions as an ad broker. However, in other
embodiments, ads may be served directly to user interfaces from
remote ad servers. In a case such as this, user preference lists
would be propagated directly to such ad servers and server 51 of
FIG. 3 would be responsible only for target information generation
(generating preference lists). It is noted herein, that step 69,
73, and 75 occur during the small amount time required for a user
to login and access a personalized Web page.
[0081] At step 77, graphical ads are delivered to user interfaces
associated with wired PCs, laptops, and other appliances having
suitable capabilities for graphical display. At step 79, textual
ads are delivered to wireless users accessing their personalized
pages via cellular phones, handheld computers, two-way pagers, and
other Internet-capable wireless appliances. It is noted herein that
in some cases of display, graphical ads may be reduced somewhat for
wireless interfaces but still retain some graphical integrity.
Similarly, interfacing devices suited for more elaborate graphical
displays may still, in some cases, receive only text ads.
[0082] It will be apparent to one with skill in the art, that the
process steps illustrated in this example are exemplary only and
may be presented in a number of different orders and descriptions
representing various embodiments without departing from the spirit
and scope of the present invention. For example, in an embodiment
where ad matching to preference criteria is performed outside of
the domain of the service provider such as within the domain of ad
partners, a step would be included for delivering categorized
preference lists to various target ad servers hosted by associated
ad partners. The inventor intends that the exemplary steps
illustrated herein represent just one of many possible processes
that may define dynamic ad serving based on evolving user
preferences.
[0083] In one embodiment of the present invention, the process of
polling data sources for data that may be matched to generated
preference lists may incorporate embodiments wherein banner and
text ads are just a few of the types of advertisements that may be
routed into an interface. For example, in addition to ads normally
served by ad servers, the method and apparatus of the present
invention may be expanded to include email advertisements and other
types of instant messaging advertisements.
[0084] In still another embodiment of the present invention the
existence of a concise and packaged preference list can be utilized
in addition to search phrases or keywords entered into a
conventional search engine in order to provide a list of URLs that
most closely and match actual user preferences that may in some way
relate to the executed data search. In this case distributed
software would be required at the database locations hosted by the
entities providing the data search services. In all aspects of the
present invention, the processes of generating a preference list,
identifying and associating ads to the preference list, and
delivering the selected ads to the user on all accomplished during
the time window a user utilizes in the process of accessing and
downloading a personalized Web page. Ads presented within a
personalized Web page will evolve with regards to content at the
rate that the associated user profile evolves with regards to
content.
[0085] The method and apparatus of the present invention may be
practiced on any DPN that supports the appropriate Internet
protocols. Furthermore, there's no limit to the number of cobrand
partners, or end-users that may participate in the practice the
present invention. Therefore, the method of the present invention
should be afforded the broadest possible scope under examination.
The spirit and scope of the present invention is limited only by
the claims that follow.
* * * * *