U.S. patent application number 14/321864 was filed with the patent office on 2015-01-22 for systems and methods for recommending purchases.
The applicant listed for this patent is Pipit Interactive, Inc.. Invention is credited to Srinivasa Boppana, David Fishman.
Application Number | 20150025996 14/321864 |
Document ID | / |
Family ID | 52344341 |
Filed Date | 2015-01-22 |
United States Patent
Application |
20150025996 |
Kind Code |
A1 |
Fishman; David ; et
al. |
January 22, 2015 |
SYSTEMS AND METHODS FOR RECOMMENDING PURCHASES
Abstract
A system and method for creating and sharing e-commerce
inventory among a defined on-line community via a specialized
visualization and interactivity interface is presented. One
embodiment comprises a network system having a client-server
architecture configured for exchanging data over a network. The
data exchanges may pertain to various functions, such as on-line
purchases, etc., and aspects, such as managing social networks,
etc., associated with the network system. The network system may
include a network-based marketplace, such as an e-commerce system,
where traders or users may communicate and exchange data. A
recommendation engine scrapes metadata from items selected in an
"art-board" and extrapolates that data to recommend other items.
These recommended other items should be based-on what other users
have selected from the "art-board" and paired together, as well as
what other users have removed from their respective art boards.
Inventors: |
Fishman; David; (Los
Angeles, CA) ; Boppana; Srinivasa; (Irvine,
CA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Pipit Interactive, Inc. |
Los Angeles |
CA |
US |
|
|
Family ID: |
52344341 |
Appl. No.: |
14/321864 |
Filed: |
July 2, 2014 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
61847656 |
Jul 18, 2013 |
|
|
|
61847890 |
Jul 18, 2013 |
|
|
|
Current U.S.
Class: |
705/26.7 |
Current CPC
Class: |
G06Q 50/01 20130101;
G06Q 30/0641 20130101; G06Q 30/0603 20130101; G06Q 30/0631
20130101; G06Q 30/0643 20130101; G06Q 30/0627 20130101 |
Class at
Publication: |
705/26.7 |
International
Class: |
G06Q 30/06 20060101
G06Q030/06 |
Claims
1. A method for performing e-commerce, comprising: activating an
art board from a network-based marketplace; placing one or more
items onto the art board; recommending additional items to place on
the art board; inviting users to interact with the art board; and
collaborating regarding art board items.
2. The method of claim 1, comprising inviting users selected from a
buddy list.
3. The method of claim 1, wherein the collaborating comprises voice
chatting, video chatting, instant messaging, or text messaging.
4. The method of claim 1, comprising examining reviews, ratings,
reputations, and recommendations.
5. The method of claim 1, comprising displaying a toolbar on the
art board and used to inviting of users and placing of items onto
the art board.
6. The method of claim 1, comprising generating a user interface
written in HTML5 for a web-based storefront with a recommendation
engine and transactional capabilities built-in), as well as a
mobile rich-media ad unit with very similar functionality of the
full version of the product (the web-based store front), except in
a smaller scale with more targeted options
7. The method of claim 1, comprising scraping metadata from items
selected in the art-board and extrapolating data to recommend other
items.
8. The method of claim 1, wherein the recommended other items are
based-on what other users have selected from the art-board and
paired together, and what other users have removed from their
respective art boards.
9. The method of claim 1, comprising recommending a similar
art-board or an alternative art-board based on product
attribute(s), visitor behavioral attribute(s), content
attribute(s), associated attributes.
10. The method of claim 1, comprising performing real time and
aggregated recommendations, building the recommendation engine by
mining data in real-time to provide relevant results and generating
real-time metrics based upon user patterns.
11. A network system, comprising: a processor; a database coupled
to the processor; and a data storage device coupled to the
processor containing code for: activating an art board from a
network-based marketplace; placing one or more items onto the art
board; recommending additional items to place on the art board;
inviting users to interact with the art board; and collaborating
regarding art board items.
12. The system of claim 11, comprising code for inviting users
selected from a buddy list.
13. The system of claim 11, wherein the collaborating comprises
voice chatting, video chatting, instant messaging, or text
messaging.
14. The system of claim 11, comprising code for examining reviews,
ratings, reputations, and recommendations.
15. The system of claim 11, comprising code for displaying a
toolbar on the art board and used to inviting of users and placing
of items onto the art board.
16. The system of claim 11, comprising code for generating a user
interface written in HTML5 for a web-based storefront with a
recommendation engine and transactional capabilities built-in), as
well as a mobile rich-media ad unit with very similar functionality
of the full version of the product (the web-based store front),
except in a smaller scale with more targeted options
17. The system of claim 11, comprising code for scraping metadata
from items selected in the art-board and extrapolating data to
recommend other items.
18. The system of claim 11, wherein the recommended other items are
based-on what other users have selected from the art-board and
paired together, and what other users have removed from their
respective art boards.
19. The system of claim 11, comprising code for recommending a
similar art-board or an alternative art-board based on product
attribute(s), visitor behavioral attribute(s), content
attribute(s), associated attributes.
20. The system of claim 11, comprising code for performing real
time and aggregated recommendations, building the recommendation
engine by mining data in real-time to provide relevant results and
generating real-time metrics based upon user patterns.
Description
[0001] The present application claims priority to Provisional
Application Nos. 61/847,656 filed Jul. 18, 2013 and 61/847,890
filed Jul. 18, 2013.
[0002] The present invention is directed to a system and method for
online shopping.
[0003] In recent years, with the popularization of the Internet and
particularly of the World Wide Web, online shopping has
revolutionized the retail industry. In contrast to brick-and-mortar
malls, online shopping can be conducted from the privacy of the
customer's home. In contrast to traditional catalog shopping, the
buyer does not have to communicate the order to the retailer by
mail, facsimile, or telephone; instead, the buyer can simply point
and click to order. Also, since a human order taker does not have
to read the order form or take the order over the telephone, the
order can be fulfilled quickly and accurately.
[0004] However, online shopping also has the disadvantage that the
buyer cannot physically inspect the item. While that disadvantage
is minor for bookstores, it is a major problem for apparel
retailers, since customers prefer to try on apparel before
buying.
[0005] To overcome that disadvantage, various techniques for
virtual modeling of apparel, particularly eyewear, have been
developed. An illustrative example of such a technique is disclosed
in U.S. Pat. No. 5,983,201 to Fay. The online retailer obtains
digital images of the customer's head and face to obtain size and
image data. Later, the customer can visit the online retailer's Web
site from any location, such as the customer's home, to view
various kinds of eyeglasses. The online retailer's server generates
images of the customer with the eyeglasses resized to fit the
customer's head to show how the customer would look in each kind of
eyeglasses.
[0006] Apparel shopping is a social event. Many customers do not
simply wish to see for themselves how they would look in a
particular item of apparel; instead, they bring along friends or
family members and solicit those friends' or family members'
opinions before making a buying decision. Shoppers may also solicit
the opinions of store clerks or of complete strangers. It is
difficult to do any of those things in front of a computer.
Furthermore, trips to brick-and-mortar shopping malls have a social
role that online shopping has not yet duplicated.
[0007] It is also known in the art to allow potential buyers to
exchange information about items over the Internet. Such
information exchanges typically take the form of non-real-time
message boards such as those on Deja.com, or the reader reviews of
Amazon.com. The use of chat rooms to let potential buyers exchange
information is taught by U.S. Pat. No. 6,041,311 to Chislenko et
al, U.S. Pat. No. 6,049,777 to Sheena et al and U.S. Pat. No.
6,058,379 to Odom et al. However, such information exchanges do not
overcome the above-noted problems with Fay and similar techniques.
U.S. Pat. No. 6,901,379 discloses a system that allows a user to
browse an online retailer's Web site or a mirror site and select an
item and model the item online by having a server generate a
digital image of the user wearing the item. If the user is still
unsure as to whether to buy the item, the user can enter an online
chat room in which the online modeling image is displayed to other
users. The user can then receive the other users' feedback before
deciding whether to buy the item. In a second embodiment, multiple
online modeling images are generated to provide the user with a
customized catalog, which can be of items for a single merchant or
multiple merchants.
[0008] U.S. Pat. No. 7,949,659 discloses systems for selecting
items to recommend to a user. The system includes a recommendation
engine with a plurality of recommenders, and each recommender
identifies a different type of reason for recommending items. In
one embodiment, each recommender retrieves item preference data and
generates candidate recommendations responsive to a subset of that
data. The recommenders also score the candidate recommendations. In
certain embodiments, a normalization engine normalizes the scores
of the candidate recommendations provided by each recommender. A
candidate selector selects at least a portion of the candidate
recommendations based on the normalized scores to provide as
recommendations to the user. The candidate selector also outputs
the recommendations with associated reasons for recommending the
items.
[0009] U.S. Pat. No. 8,170,919, issued to the assignee of the
instant invention, discloses an inventive system and method for
collaborative commerce that includes activating an art board,
placing items onto the art board, inviting users to interact with
the art board, and collaborating with the invited users. Additional
features and functions include purchasing items shown on the art
board, including by placing the items in a shopping cart, using
e-mail, text messaging, and instant messaging to invite users, who
may be chosen from a buddy list. Collaborating can be performed
using voice chatting, video chatting, instant messaging, and text
messaging, and includes examining reviews, ratings, reputations,
and recommendations, and also includes displaying details regarding
the items. In addition, reports comprising information regarding
the items can be generated. A toolbar can be located on the art
board and used to initiate inviting of users and placing of items
onto the art board.
SUMMARY
[0010] A system and method for creating and sharing e-commerce
product and/or content inventory among a defined on-line community
via a specialized visualization and interactivity interface is
presented. One embodiment comprises a network system having a
client-server/Web services architecture configured for exchanging
data over a network. The data exchanges may pertain to various
functions, such as on-line purchases, etc., and aspects, such as
managing social networks, etc., associated with the network system.
The network system may include a network-based marketplace, such as
an e-commerce system, where traders or users may create, consume,
communicate and exchange data. The Recommendation engine consumes
the attributes from a website, specific section of the site,
channel, product, content, and visitors behavioral data including
but not limited to Geographic, navigational, website, product,
content, search keywords (onsite or offsite) attributes and then
recommend art boards or collections based on one or more of those
attributes.
[0011] Recommendation system can be available on any website by
adding the webservices logic to communicate with the recommendation
engine. Recommendation modules can displayed as a Display Ad module
on external web and mobile sites/apps. Recommendation engine can be
accessible on any device that can communicate via web or internet
that can communicate via web services.
[0012] A recommendation engine scrapes metadata from items selected
in an "art-board" and extrapolates that data to recommend other
items. These recommended other items should be based-on what other
users have selected from the "art-board" and paired together, as
well as what other users have removed from their respective art
boards. A recommendation engine also scrapes metadata [0013] a.
from art boards and extrapolates that data to recommend other art
boards. These recommended art boards may be based-on what other
users have liked, viewed, shared from the "art-board" and/or paired
together, as well as what other users have removed from their
respective likes or shares. [0014] b. from a given product
attribute(s) (for e.g., brand name, or product category, or price,
etc.,), visitor behavioral attribute(s) (colors, sections of a
website, or geographic information), content attribute(s) (names,
titles, etc.,) and/or associated attributes (search keywords,
etc.,) and recommend art boards that are relavant or similar, or
provide alterantive art boards.
[0015] In one implementation, a web-based product provides ways to
access the recommendation engine services using web services and
leveraging a user interface that's written in HTML5/web/native
mobile app, which can be both a web-based storefront (with the
recommendation engine and transactional capabilities built-in), as
well as a mobile rich-media ad unit with very similar functionality
of the full version of the product (the web-based store front),
except in a smaller scale with more targeted options (based on the
demographic the mobile ad is displayed-to). This product will
function as a personal shopping recommendation engine, and will, at
some point, also be integrated in the brick-and-mortar/physical
retail shopping experience
[0016] In one embodiment, the system can recommend products that
others with similar preferences have "liked" or "disliked,"
allowing users to create custom curated collections that are
influenced by the process (essentially, influenced by other users
of the system as well as the system's pairing for tags included in
the metadata).
[0017] The present invention advantageously provides a system and
method for aggregating product information onto a visualization
board from products initially resident on a network-based
marketplace, such as a website, or a mobile app or an e-commerce
web site. The products and their aggregated product information can
be displayed on the visualization board in a manner enabling
multiple users to interact and view groups of products from single
or multiple websites, while the users are engaging, chatting and
interacting on-line with each other and, if desired, traversing
among web sites.
[0018] The inventive system and method comprises activating an art
board, automatically recommending items onto the art board from a
user's recently viewed items palette, inviting users to interact
with the art board and other users' recently viewed items palettes,
and collaborating with the invited users. An art board can be
activated whereas user's recently viewed items can be made public
to any networked user of the system. Additional features and
functions include purchasing items shown on the art board,
including by placing the items in a shopping cart, using e-mail,
text messaging, and instant messaging to invite users, who may be
chosen from a buddy list. Collaborating can be performed using
voice chatting, video chatting, instant messaging, and text
messaging, and includes examining reviews, ratings, reputations,
and recommendations, and also includes displaying details regarding
the items. In addition, reports comprising information regarding
the items can be generated. A toolbar can be located on the art
board and used to initiate inviting of users, creation of new art
boards, and copy and/or save of art boards. The recently viewed
items palette can be located on the art board and shared with other
users of the system based on permissions. The system enables
creating and sharing of product, content, or e-commerce inventory
among a defined on-line community via a specialized visualization
and interactivity interface. One embodiment comprises a network
system having a webservices / client-server architecture configured
for exchanging data over a network. The data exchanges may pertain
to various functions, such as on-line purchases, etc., and aspects,
such as managing social networks, etc., associated with the network
system. The network system may include a network-based marketplace,
such as an e-commerce system, where traders or users may
communicate and exchange data.
BRIEF DESCRIPTION OF THE DRAWINGS
[0019] The invention is further described in the detailed
description that follows, by reference to the noted drawings by way
of non-limiting illustrative embodiments of the invention, in which
like reference numerals represent similar parts throughout the
drawings. As should be understood, however, the invention is not
limited to the precise arrangements and instrumentalities shown. In
the drawings:
[0020] FIG. 1 shows an exemplary process for recommending items on
an interactive art-board.
[0021] FIG. 2 shows an exemplary art-board system for creating and
sharing e-commerce inventory among a defined on-line community
through a specialized visualization and interactive interface.
[0022] FIG. 3 shows an exemplary recommended Art-Boards for a
Spring Shopper (visitor behavior) or based on a "Fashion"
keyword.
[0023] FIG. 4 shows a mockup of potential use of "Recommendation"
module consuming the visitors data (in this case "Style").
DETAILED DESCRIPTIONS
[0024] FIG. 1 shows a flow diagram of an exemplary method of the
invention. In step S1, a user or host activates an art board 10
(FIG. 2) from a network-based marketplace 36. In step S2, one or
more items are placed onto the art board 10. In step S3, a
recommendation engine is used to recommend additional items to
place on the art board 10. Next, in S4, the host invites other
users to interact with the art board 10. This invitation can be
made via e-mail, SMS, IM or other electronic means. In step S5,
users collaborate regarding art board items. This collaboration can
include chatting, instant messaging, adding items to the art board,
etc. In step S6, one or more users, including the host, can
purchase items shown on the art board 10. In one embodiment, items
are purchased by moving them to an electronic shopping cart, paying
electronically, and having the items delivered to the purchaser.
Optional step S7 can produce reports, such as statistical data
relating to products and/or user preferences in accordance with the
items on the art board.
[0025] Steps S2 through S6 can be performed in any order and each
step can be performed more than one time. For example, the host can
invite two friends, step S4, the three users can collaborate, step
S5, one or more of the users can place one or more items onto the
art board, step S3, a user can purchase an item S6, etc.
[0026] In one implementation, the process includes activating an
art board, automatically recommending and placing items onto the
art board, inviting users to interact with the art board, and
collaborating with the invited users. Additional features and
functions include purchasing items shown on the art board,
including by placing the items in a shopping cart, using e-mail,
text messaging, and instant messaging to invite users, who may be
chosen from a buddy list. Collaborating can be performed using
voice chatting, video chatting, instant messaging, and text
messaging, and includes examining reviews, ratings, reputations,
and recommendations, and also includes displaying details regarding
the items. In addition, reports comprising information regarding
the items can be generated. A toolbar can be located on the art
board and used to initiate inviting of users and placing of items
onto the art board.
[0027] Next, the recommendation engine operation is discussed. In
one embodiment, a web-based product with a user interface that's
written in HTML5, which can be both a web-based storefront (with
the recommendation engine and transactional capabilities built-in),
as well as a mobile rich-media ad unit with very similar
functionality of the full version of the product (the web-based
store front), except in a smaller scale with more targeted options
(based on the demographic the mobile ad is displayed-to). This
product will function as a personal shopping recommendation engine,
and will, at some point, also be integrated in the
brick-and-mortar/physical retail shopping experience.
[0028] The recommendation engine scrapes metadata from items
selected in a "art-board" and extrapolates that data to recommend
other items. These recommended other items should be based-on what
other users have selected from the "art-board" and paired together,
as well as what other users have removed from their respective art
boards.
[0029] Essentially, this process is analogous to Pandora's music
genome project, recommending music that others with similar
preferences have "liked" or "disliked," allowing users to create
custom curated collections that are influenced by the process
(essentially, influenced by other users of the system as well as
the system's pairing for tags included in the metadata).
[0030] The recommendations processes operate by attempting to match
users to other users having similar behaviors or interests. For
example, once Users A and B have been matched, items favorably
sampled by User A but not yet sampled by User B may be recommended
to User B. In contrast, content-based recommendation systems seek
to identify items having content (e.g., text) that is similar to
the content of items selected by the user. Other recommendation
systems use item-to-item similarity mappings to generate the
personalized recommendations. The item-to-item mappings may be
generated periodically based on computer-detected correlations
between the item purchases, item viewing events, or other types of
item selection actions of a population of users. Once generated, a
dataset of item-to-item mappings may be used to identify and
recommend items similar to those already "known" to be of interest
to the user. In one embodiment, the system can be built by: [0031]
1. Writing a process (or set of processes) that conforms-to
(utilizes) the Purchlive system backend (with its
tags/categorization, as well as content population feature). [0032]
2. Building a new backend with a similar feature-set to the current
Purchlive system (namely, one that has tags/categorization, as well
as content population feature) that integrates an process (or set
of process) that accomplish the purpose (as listed in the first
paragraph of this document).
[0033] When building the recommendation engine, several factors
must be taken into account. The engine can consider factors such
as:
[0034] Data Available for process
[0035] Real time vs. Aggregated recommendations
[0036] Weighting suggestions (process definition)
[0037] External data usage
[0038] Third party recommendation engines
One embodiment uses available data such as, but not limited to:
[0039] Product Side Data--uploaded into the system by customer
[0040] Title [0041] Description [0042] Current Price (or Sale
Price) [0043] MSRP (Non-Sale Price) [0044] Keywords/tags
[0045] App Data--produced when users take action [0046] Products
Viewed [0047] Products Added to Artboard [0048] Products Saved in
Artboard [0049] Products Shared within Artboard [0050] Products
Commented upon (Must integrate with FB)
[0051] Visit Data--this data can be obtained from Google Analytics
[0052] Customers viewing, adding, saving, sharing products [0053]
Interactions per Product [0054] Overall Interactions [0055] (not
real-time) Geographic Location
[0056] Product Sales Data--Customer Side [0057] Orders [0058] Sales
[0059] Revenue [0060] Cost of Goods
[0061] In an embodiment, the recommender analyzes a subset of the
item preference data to identify items as candidate recommendations
for recommending to a user. Each recommender also identifies one or
more reasons for recommending the items. As discussed below,
different recommenders may use different types of item preference
data than others to select candidate items to recommend. Different
recommenders may also provide different types of reasons for
recommending items.
[0062] For example, a particular recommender might retrieve the
user's purchase history data. Using this data, the recommender can
find items owned by the user that are part of a series. A series
might include, for instance, books in a trilogy, movies and their
sequels, or all albums by a musician. If the user has purchased
fewer than all the items in the series, the recommender might
select the remaining items as candidate recommendations and provide
a reason such as, "this item is recommended because you purchased
items A and B, and this item would complete your series."
Advantageously, this reason can be more compelling than a reason
such as "because you purchased items A and B, and this item is
similar." Users may therefore be more inclined to trust the reasons
provided by the recommenders.
[0063] As another example, a recommender might obtain data about a
user's friends. This friends data might include information on the
friends' birthdays, their wish lists, and their purchase histories.
Using this data, a recommender might suggest gifts that could be
bought for a friend's upcoming birthday and provide a reason such
as "this item is recommended because your friend John's birthday is
on July 5th, and this item is on his wish list." Provided with such
a reason, the user might be more inclined to buy the item.
[0064] Many other examples of item preference data may be used by
the recommenders to generate candidate recommendations and
corresponding reasons. For instance, browse history data (e.g.,
data on user searches, clicks, and the like) may be used to provide
a recommendation with the reason, "because this item is similar to
an item you searched for." Purchase history data and/or wish list
data might be used to provide a recommendation with the reason,
"because this item might be interesting to an early adopter such as
you." Browse history data on a browse node of interest to the user
(e.g., a category browsed by the user) might be used to provide a
recommendation with the reason, "because this item is a top seller
in one of your favorite interest areas." Various other forms of
item preference data may be used to provide recommendations with
reasons such as "because you recently moved," "because you bought
an item that may need replacing," "because most people upgrade
their shoes after two years," or the like.
[0065] Multiple reasons may be provided by a single recommender, or
multiple recommenders may each provide the same candidate
recommendation along with a different reason for that
recommendation. For instance, several recommenders may be used to
recommend a particular pant because 1) a user recently rated
several pants, 2) this is the best-selling movie in the pant
category, and 3) this pant was nominated for two awards. Using
multiple reasons may provide further motivation to the user to view
or buy an item.
[0066] The system can perform real time as well as aggregated
recommendations. When building the recommendation engine there will
be data that can be mined in real-time to provide relevant results.
These real-time metrics will be constantly available based upon
user patterns. Other metrics will come from outside sources and may
be combined into the process on a batch process. The system can
include a mobile-ad unit which can perform collecting real-time
metrics within ads.
[0067] With limited data at the start, the recommendations and the
data needed to make good ones may take some time. In order to
assure that products start to get an even chance at rankings
(products towards back of stack browser may never get the right
chance), we should start to randomize the display of products
within the creation process. Currently the same shoes, skirts, etc.
are being displayed up front which will skew results.
[0068] In one embodiment of a Real-time Weighted process, due to
the lack of data upon launch, a linear process will be built that
simply provides point values to products. This will determine both
placement of products and order in which collections are displayed
to users. As data becomes more available to each product and its
related entities a stronger and more sloped process can be
utilized.
[0069] Next, Product Weighting can be done. Each Product will
record the following numeric values any time a product is
interacted with. The higher the score, the higher the chances that
products will be displayed to users. This weighting happens
pre-user activity. Each product view will also be recorded, and
divided into the current weighting scale. This will allow newer
products introduced to still grow relevance. For example, the
product can be scored as follows:
[0070] 1 pt (simple interaction)
[0071] 3 pt (adding interaction)
[0072] 5 pt (sharing interaction)
[0073] 10 pt (buying interaction)
[0074] The system then performs Product Relationship weighting. In
one embodiment, a set of tables will be managed recording each
product and how it was used with other products. This will be
summed and influence the display of products after a user starts to
show interest.
[0075] 1 pt (added Product A with Product B)
[0076] 2 pt (user has selected Product A and Product B but not in
same outfit)
[0077] 3 pt (outfit saved with Product A and Product B)
[0078] 5 pt (outfit shared with Product A and Product B)
[0079] Collection Relationship weighting is performed next in one
embodiment. This additional weighting will be done for collections
that are very similar to the Product weighting scale used above.
This weighting occurs pre and post user interaction. Pre post
allows for proper collection display and ranking while post use
interaction weights can be used to provide "other interested" looks
or collections. As with Product weights, the sum of the values will
be divided by the number of impressions a collection receives.
[0080] 1 pt (browsed collection)
[0081] 3 pt (shopped collection)
[0082] 5 pt (edited collection)
[0083] Third Party Recommendation Engines may be applied to the
process. Once such engine is Brilliance, an affordable solution
with little integration needs. With over 20 recommendations
including products bought after matches, customers also liked,
bundled products, and more. Another engine is Strands--with a very
involved integration. The system can incorporate Strands in many
areas and take advantage of reviews and ratings from the system
users, and there is also a iPhone SDK available to incorporate
recommendations into an app. Strands allows for several built-in as
well as configurable recommendation types. Yet another
recommendation engine is Certona, a popular of all recommendation
engines, as well as the most robust and customized. Rich
relevance--{rr} is also used on some of the internee's largest
sites including O.com, WalMart and Sears.
[0084] In one implementation, the recommender engine can: [0085]
categorize customer tags (summer shoes), comments from art board
[0086] categorize likes/dislikes/ratings/sharing option [0087]
categorize based on the art board item's parent product attributes,
including product names, categories, manufacturers, color,
size=material, price, sale price, campaign and other attributes
from product catalog Ad/targeting options (Able to present ad
units/runway)
[0088] Recommendation engine [0089] Merchandising override rules
with predefined list of items based on [0090] specific category
[0091] specific campaign [0092] attributes (based on price, sale
price, inventory, campaign, color, brand, popularity, size);--
Dynamic--based on
[0093] Also liked [0094] Also viewed [0095] Tags [0096] popular for
each category [0097] product attributes
Publishers:
[0097] [0098] capture the site URL, page meta data where the art
board is created, accessed Analytics can perform the following:
Track site information for e.g., where the art board is created,
and other site attribute=(portal, social media, blogs, etc.,) Track
technology information Track interactions (click, slide, add,
remove, share=like, save, views)
[0099] FIG. 2 shows a schematic illustration of an exemplary
embodiment of the present invention. As shown in FIG. 2, when a
visualization board 10 is activated, a new user interface layer is
created above the normal user interface screens associated with a
network-based marketplace 36. This visualization layer visually
appears as a floating window above the normal web interface
associated with the network-based marketplace 36. In one
embodiment, FLEX Data Services from Adobe Corporation can be used
to support the presentation layer of the visualization board 10 and
to provide ways to send and load data to and from server-side
components without requiring the client to reload the view of the
underlying network application layers. In one embodiment, the
visualization board 10 is PurchLive, available at
www.purchlive.com. PurchLive is an advanced user-generated content
engine, combined with a powerful analytics engine, that allows a
customer to use existing content in new ways and generates fresh
insights into consumer behavior. By adding a new dimension to the
web site, PurchLive leverages the user's existing investment and
creates a valuable new viral marketing channel.
[0100] FIG. 3 shows an exemplary recommended Art-Boards for a
Spring Shopper (visitor behavior) or based on a "Fashion" keyword,
while FIG. 4 shows a mockup of potential use of "Recommendation"
module consuming the visitors data (in this case "Style"). From a
given product attribute(s) (for e.g., brand name, or product
category, or price, etc.,), visitor behavioral attribute(s)
(colors, sections of a website, or geographic information), content
attribute(s) (names, titles, etc.,) and/or associated attributes
(search keywords, etc.,) and recommend art boards that are relevant
or similar, or provide alternative art boards. As shown in FIG. 3,
another application of the recommendation engine is shown. The
system can show recommended art boards based on a given product or
a product category page attributes or other attributes (say search
keyword attributes). In this embodiment, the art board will
dynamically change based on the product/content/other attributes
the user are providing to the recommendation engine. For example if
the user is on a "Style" article page, the "recommendation" module
will show dynamically/real-time relevant artboards (outfits) that
match with "Style" or "fashion" or related Art boards.
[0101] In one embodiment, the recommender engine automatically
suggests items available for purchase that best matches the user's
interest and places these items onto the visualization board 10.
Other users can also place items onto the visualization board 10.
At all times, all users see the same visualization board 10,
including users invited into the session after its initialization.
Any product added or removed, or any Cartesian movement of products
on the visualization board 10, is automatically synchronized to the
visualization boards seen by other users. This is because any
manipulation of the visualization board 10 by any one user is
automatically annotated into a server, e.g. the network-based
provider, that synchronizes the visualization board 10 of all other
users members of that session. This also applies to other data,
such as annotations, product information and metadata, that any
user could add.
[0102] The visualization board 10 combined with the network and
marketplace applications can include one or more applications which
support the network-based marketplace, and can generate and
maintain relationships between products, community groups and their
members' rules and roles, and transactions that may be associated
with the network-based marketplace shopping cart including the
products purchased through it. The associated relationships may
include distribution parameters, e.g., roles and rules pertaining
to the item list and associated community group(s), reviews and
recommendations pertaining to the items of the item list, item
attributes like model and manufacturer, or service provider of a
particular item, item status, e.g., purchased, etc. Additionally,
the various applications may support social networking functions,
including building and maintaining the community groups created by
a user, relating one or more item lists to selected community
groups, and providing a shared electronic shopping cart for the
community groups to purchase items from the shared item list.
[0103] On-line store or e-commerce applications may allow sellers
to group their listings, e.g., goods and/or services, in the
visualization boards 10 within a "virtual" store, which may be
branded and otherwise personalized by and for the sellers. Such
virtual storyboards 10 may also offer promotions, incentives and
features that are specific and personalized to a relevant seller.
In one embodiment, the listings and/or transactions associated with
the virtual storyboards and their features may be provided to one
or more community groups having an existing relationship with the
item list creator. An existing relationship or association may
include a friend or family relationship, a transactional
relationship, e.g., prior sales with user, or an overall network
community relationship, e.g., buyers historical transaction rating.
Reputation applications may allow parties that transact utilizing
the network-based marketplace 36 and the storyboards 10 to
establish, build and maintain reputations, which may be made
available and published to potential trading partners.
[0104] A number of fraud prevention applications may implement
various fraud detection and prevention mechanisms to reduce the
occurrence of fraud within the marketplace. In one embodiment, the
fraud prevention applications may monitor activities of each user
within the community group. For example, the item list creator may
want to be informed if a member of the community group adds items
to the virtual storyboard or changes shipping information, provided
the member had the necessary permissions. In various embodiments,
whether to monitor and the level of monitoring may depend upon the
relationship to the item list creator. For example, an indirect
relationship may be more heavily monitored than a direct
relationship.
[0105] Messaging applications may be used for the generation and
delivery of messages to users of the network-based marketplace 36.
Messages can, for example, advise the visualization board creator
and members of the community groups associated with an item list of
the status of the various items on the list, e.g., already
purchased, etc. In one embodiment, the messaging applications may
be used in conjunction with the social networking applications to
provide promotional and/or marketing information to the community
members associated with the item list to assist them in finding and
purchasing items on the visualization board 10.
[0106] A reporting application connected with the virtual
storyboard 10 can compile statistical data relating to the
products, selection, choices, and/or preferences of users with
respect to selecting products and/or combinations. A ranking system
could be created whereby such information is compiled statistically
and made available to merchants for trend analysis. Additionally
such information could be combined with "recommendation engines" to
suggest products automatically or manually. In one embodiment, such
recommendation engine could include a collaborative filtering
engine that catalogs and indexes similar users with their choices
of products and recommends the choices of one similar user to the
others.
[0107] A user table may contain a record for each registered user
of the network-based marketplace, and may include identifier,
address and financial instrument information pertaining to each
such registered user. In one embodiment, a user operates as an item
visualization board creator or a member of a community group,
including associated operations pertaining to the rules and roles,
created by the visualization board creator. A user may also operate
as a seller, a buyer, or both, within the network-based
marketplace. The tables may also include a visualization board
table that maintains listing or item records for goods and/or
services created by a visualization board creator. In one
embodiment, the visualization board is created for sharing with a
community group defined, at least in part, by the visualization
board creator.
[0108] Furthermore, each listing or item record within the
visualization board table may be linked to one or more electronic
shopping cart records within a electronic shopping cart table and
to one or more user records within the user table and/or a vendor
table, to associate a seller or vendor and one or more actual or
potential buyers from the community group with each visualization
board.
[0109] A transaction table may contain a record for each
transaction pertaining to items or listings for which the user
defined community group rules and roles pertain to one or more
items of the visualization board. For example, the visualization
board creator may not want a member of a community group to be able
to view, purchase, edit, etc., any or all of the items in the
visualization board. In another example, rules may include an
ability to purchase an item on the list, purchase one or more items
using the creator's account, add to the visualization board,
etc.
[0110] Additionally, the visualization board creator may want to
assign roles to an entity within the community group. For example,
roles may include a buyer, a reviewer, an administrator, etc.
Accordingly, a rules applications and a roles applications may be
used in conjunction with social networking applications to
customize the visualization board to be shared within one or more
community groups.
[0111] The relationship or association between the visualization
board creator (user) and the members of the one or more community
groups may be a direct relationship or an indirect relationship. An
example of a direct relationship may be a sister, a friend, or a
trusted associate user, while the indirect relationship may be a
secondary entity brought in by a direct relationship. The web
servers can access one or more additional repositories of user
data. Because a group of individuals can share an account, a given
"user" may include multiple individuals (e.g., two family members
that share a computer). The data stored for each user may include
one or more of the following types of information (among other
things) that can be used to generate recommendations in accordance
with the engine: (a) the user's purchase history, including dates
of purchase, (b) a history of items recently viewed by the user,
(c) the user's item ratings profile, if any, and (d) items tagged
by the user. Various other types of user information, such as wish
list/registry contents, email addresses, shipping addresses,
shopping cart contents, and browse (e.g., clickstream) histories,
may additionally be stored.
[0112] The network system also includes a network-based provider
having a data exchange platform, such as an art board, to provide
server-side functionality via a network, e.g., the
[0113] Internet, to one or more clients, including users that may
utilize the network system through the network-based provider to
exchange data over the network. The data exchange may include
transactions such as receiving and processing data from a multitude
of users. The data may include, but is not limited to, shared
recently viewed products, product and service reviews, product,
service, manufacture, and vendor recommendations, product and
service listings, auction bids, feedback, etc.
[0114] In an exemplary embodiment, the network-based marketplace,
the network-based provider including the data exchange platform, an
application program interface (API) server, and a web server are
coupled to, and provide programmatic and web interfaces
respectively to, one or more application servers. The application
servers host one or more networking applications and marketplace
applications. The applications servers, in turn, are coupled to one
or more database servers that facilitate access to one or more
databases. The marketplace application may provide a number of
marketplace functions and services, e.g., listing, payment, etc.,
to users that access the network-based marketplace.
[0115] This inventive system also embodies the notion of a third
party application, executing on a third party server machine, as
having programmatic access to the network-based marketplace via the
programmatic interface provided by the API server. For example, the
third party application may, utilizing information retrieved from
the network-based marketplace, support one or more features or
functions on a website hosted by the third party. The third party
website may, for example, provide one or more networking,
marketplace or payment functions that are supported by the relevant
applications of the network-based marketplace. Under such
embodiments, multiple network and marketplace applications,
respectively, could be part of the network-based marketplace.
[0116] Various other applications, separate or as part of the
network-based marketplace, may support social networking functions.
These could include allowing the user to create groups of other
users, affiliates, and lists of friends, and to facilitate various
group communications to those lists and users, including
distributing products in the network-based marketplace. While the
social networking applications and the marketplace applications are
discussed here as joined to form part of the network-based
marketplace, in alternative embodiments, the networking
applications may form part of a social networking service that is
separate and distinct from the marketplace.
[0117] The various components of the web site system may run, for
example, on one or more servers (not shown). In one embodiment,
various components in or communicating with the recommendations
service are replicated across multiple machines to accommodate
heavy loads.
[0118] Each of the processes and algorithms described above may be
embodied in, and fully automated by, code modules executed by one
or more computers or computer processors. The code modules may be
stored on any type of computer-readable medium or computer storage
device. The processes and algorithms may also be implemented
partially or wholly in application-specific circuitry. The results
of the disclosed processes and process steps may be stored,
persistently or otherwise, in any type of computer storage.
[0119] The various features and processes described above may be
used independently of one another, or may be combined in various
ways. All possible combinations and subcombinations are intended to
fall within the scope of this disclosure. In addition, certain
method or process steps may be omitted in some implementations.
[0120] Although this disclosure has been described in terms of
certain example embodiments and applications, other embodiments and
applications that are apparent to those of ordinary skill in the
art, including embodiments and applications that do not provide all
of the benefits described herein, are also within the scope of this
disclosure. The scope of the inventions is defined only by the
claims, which are intended to be construed without reference to any
definitions that may be explicitly or implicitly included in any of
the incorporated-by-reference materials.
* * * * *
References