Socially collaborative filtering for providing recommended content to a website for presentation to an identified user

• US 8,914,367 B2
• Filed: 12/15/2008
• Issued: 12/16/2014
• Est. Priority Date: 11/29/2007
• Status: Active Grant

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Inventors: Toebes, John, Kuhlke, Matthew, Scheinman, Ari Daniel
• Primary Examiner(s): Btit, Jacob F
• Assistant Examiner(s): TRAN, BAO G
• Application Number:
  US12/334,910
• Publication Number:
  US 20100153411A1
• Time in Patent Office: 2,192 Days
• Field of Search: 707/999.104, 707/999.107, 707/751, 707/944, 707/723, 707/728, 707/748, 707/798, 707/737, 707/754, 715/234, 715/745, 715/719, 709/203, 709/217
• US Class Current: 707/737
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

In one embodiment, a method comprises receiving, by a website server device providing a website service, a request from an identified user of a user device, the request requesting network content provided within the website service; determining by the website server device an absence of new network content within the website service relative to a last prior access by the identified user to the website service; and obtaining, by the website server device for presentation to the identified user within the website service, an ordered list of network items most likely to be preferred by the identified user.

Claims

1. A method comprising:

receiving, by a website server device providing a website service, a request from an identified user of a user device, the request requesting network content provided within the website service;

determining by the website server device an absence of new network content within the website service relative to a last prior access by the identified user to the website service; and

obtaining, by the website server device for presentation to the identified user within the website service, an ordered list of network items most likely to be preferred by the identified user, including sending by the website server device a recommendation request to a recommendation server via a network in response to the determined absence of new network content, and receiving the ordered list from the recommendation server via the network, wherein the ordered list is generated based on;

identifying user selection preferences of the identified user based on an accumulation of user selection inputs executed by the identified user relative to input options presented to the user and identifying respective available network items, the user selection inputs including selections of a first portion of the network content within the website service and having been consumed by the user,

determining first network users having a highest correlation of shared interests with the identified user based on identifying preferred network items from the available network items and having highest relative item affinity values generated for the identified user, and determining the first network users as providing highest relative user affinity values for the preferred network items,

identifying, as personally interesting content for the identified user, the preferred network items for each of the first network users based on the respective user selection preferences, and

filtering the personally interesting content for the identified user relative to socially related content, the socially related content determined based on identifying second network users having the highest relative user affinity values toward the network content specified in the request, and identifying second preferred network items for each of the second network users based on the respective user selection preferences;

the filtering of the personally interesting content for the identified user relative to socially related content based on generating an ordered list of content related to the website service, the ordered list of content generated based on an ordered sorting of the socially related content with contextually related content, the contextually related content determined by the website server device identifying a second portion of the network content within the website service that has not been consumed by the user, wherein the ordered list of content is ordered based on a determined convergence of relationships between identified network items in the socially related content and the contextually related content;

wherein the determined convergence of relationships is identified based on;

identifying third network users having the highest relative user affinity values toward the second portion of the network content, and identifying third preferred network items for each of the third network users based on the respective user selection preferences, and

identifying an intersection of intersecting network items between the third preferred network items and the socially related content, and determining a social correlation between the intersecting network items and at least one of the contextually related content.

2. The method of claim 1, wherein the determined convergence is determined based on assigning, to at least one network item of the second portion, a corresponding affinity value based on a corresponding prescribed similarity relationship assigned to the corresponding network item relative to another network item in the socially related content, and based on a determined absence of any user affinity value toward the at least one network item of the second portion.

3. The method of claim 1, wherein the recommendation request identifies the identified user, the request requesting network content, and the contextually related content.

4. The method of claim 1, wherein the ordered list of network items is distinct from any of the network content having been provided within the website service.

5. A method comprising:

receiving, by a server device via a network, a recommendation request from a website server device having sent the recommendation request in response to a determined absence of new network content within a website service provided by the website server device, the recommendation request specifying an identified user having sent a request, via a user device and the network, to the website service, the request requesting network content that is provided within the website service;

generating by the server device an ordered list of network items most likely to be preferred by the identified user based on the network content requested by the user; and

sending the ordered list by the server device to the website server device for presentation within the website service of the ordered list to the identified user, wherein the ordered list is generated based on;

identifying user selection preferences of the identified user based on an accumulation of user selection inputs executed by the identified user relative to input options presented to the user and identifying respective available network items, the user selection inputs including selections of a first portion of the network content within the website service and having been consumed by the user,

determining first network users having a highest correlation of shared interests with the identified user based on identifying preferred network items from the available network items and having highest relative item affinity values generated for the identified user, and determining the first network users as providing highest relative user affinity values for the preferred network items,

identifying, as personally interesting content for the identified user, the preferred network items for each of the first network users based on the respective user selection preferences, and

filtering the personally interesting content for the identified user relative to socially related content, the socially related content determined based on identifying second network users having the highest relative user affinity values toward the network content specified in the request, and identifying second preferred network items for each of the second network users based on the respective user selection preferences;

the filtering of the personally interesting content for the identified user relative to socially related content based on generating an ordered list of content related to the website service, the ordered list of content generated based on an ordered sorting of the socially related content with contextually related content, the contextually related content specified in the recommendation request and identifying a second portion of the network content within the website service that has not been consumed by the user, wherein the ordered list of content is ordered based on a determined convergence of relationships between identified network items in the socially related content and the contextually related content;

wherein the determined convergence of relationships is identified based on;

identifying third network users having the highest relative user affinity values toward the second portion of the network content, and identifying third preferred network items for each of the third network users based on the respective user selection preferences, and

identifying an intersection of intersecting network items between the third preferred network items and the socially related content, and determining a social correlation between the intersecting network items and at least one of the contextually related content.

6. The method of claim 5, wherein the determined convergence is determined based on assigning, to at least one network item of the second portion, a corresponding affinity value based on a corresponding prescribed similarity relationship assigned to the corresponding network item relative to another network item in the socially related content, and based on a determined absence of any user affinity value toward the at least one network item of the second portion.

7. The method of claim 5, wherein the ordered list of network items is distinct from any of the network content having been provided within the website service.

8. An apparatus comprising:

a network interface circuit configured for receiving, via a network, a recommendation request from a website server device having sent the recommendation request in response to a determined absence of new network content within a website service provided by the website server device, the recommendation request specifying an identified user having sent a request, via a user device and the network, to the website service, the request requesting network content that is provided within the website service; and

a processor circuit configured for generating an ordered list of network items most likely to be preferred by the identified user based on the network content requested by the user;

the network interface circuit configured for sending the ordered list to the website server device for presentation within the website service of the ordered list to the identified user, wherein the ordered list is generated based on;

identifying user selection preferences of the identified user based on an accumulation of user selection inputs executed by the identified user relative to input options presented to the user and identifying respective available network items, the user selection inputs including selections of a first portion of the network content within the website service and having been consumed by the user,

determining first network users having a highest correlation of shared interests with the identified user based on identifying preferred network items from the available network items and having highest relative item affinity values generated for the identified user, and determining the first network users as providing highest relative user affinity values for the preferred network items,

identifying, as personally interesting content for the identified user, the preferred network items for each of the first network users based on the respective user selection preferences, and

filtering the personally interesting content for the identified user relative to socially related content, the socially related content determined based on identifying second network users having the highest relative user affinity values toward the network content specified in the request, and identifying second preferred network items for each of the second network users based on the respective user selection preferences;

the filtering of the personally interesting content for the identified user relative to socially related content based on generating an ordered list of content related to the website service, the ordered list of content generated based on an ordered sorting of the socially related content with contextually related content, the contextually related content specified in the recommendation request and identifying a second portion of the network content within the website service that has not been consumed by the user, wherein the ordered list of content is ordered based on a determined convergence of relationships between identified network items in the socially related content and the contextually related content;

wherein the determined convergence of relationships is identified based on;

identifying third network users having the highest relative user affinity values toward the second portion of the network content, and identifying third preferred network items for each of the third network users based on the respective user selection preferences, and

identifying an intersection of intersecting network items between the third preferred network items and the socially related content, and determining a social correlation between the intersecting network items and at least one of the contextually related content.

9. The apparatus of claim 8, wherein the determined convergence is determined based on assigning, to at least one network item of the second portion, a corresponding affinity value based on a corresponding prescribed similarity relationship assigned to the corresponding network item relative to another network item in the socially related content, and based on a determined absence of any user affinity value toward the at least one network item of the second portion.

10. The apparatus of claim 8, wherein the ordered list of network items is distinct from any of the network content having been provided within the website service.

11. An apparatus comprising:

a network interface circuit configured for receiving a recommendation request from a website server device having sent the recommendation request in response to a determined absence of new network content within a website service provided by the website server device, the recommendation request specifying an identified user having sent a request, via a user device and the network, to the website service, the request requesting network content that is provided within the website service; and

means for generating an ordered list of network items most likely to be preferred by the identified user based on the network content requested by the user;

the network interface circuit configured for sending the ordered list to the website server device for presentation within the website service of the ordered list to the identified user, wherein the ordered list is generated based on;

identifying user selection preferences of the identified user based on an accumulation of user selection inputs executed by the identified user relative to input options presented to the user and identifying respective available network items, the user selection inputs including selections of a first portion of the network content within the website service and having been consumed by the user,

determining first network users having a highest correlation of shared interests with the identified user based on identifying preferred network items from the available network items and having highest relative item affinity values generated for the identified user, and determining the first network users as providing highest relative user affinity values for the preferred network items,

identifying, as personally interesting content for the identified user, the preferred network items for each of the first network users based on the respective user selection preferences, and

filtering the personally interesting content for the identified user relative to socially related content, the socially related content determined based on identifying second network users having the highest relative user affinity values toward the network content specified in the request, and identifying second preferred network items for each of the second network users based on the respective user selection preferences;

the filtering of the personally interesting content for the identified user relative to socially related content based on generating an ordered list of content related to the website service, the ordered list of content generated based on an ordered sorting of the socially related content with contextually related content, the contextually related content specified in the recommendation request and identifying a second portion of the network content within the website service that has not been consumed by the user, wherein the ordered list of content is ordered based on a determined convergence of relationships between identified network items in the socially related content and the contextually related content;

wherein the determined convergence of relationships is identified based on;

identifying third network users having the highest relative user affinity values toward the second portion of the network content, and identifying third preferred network items for each of the third network users based on the respective user selection preferences, and

identifying an intersection of intersecting network items between the third preferred network items and the socially related content, and determining a social correlation between the intersecting network items and at least one of the contextually related content.

12. Logic encoded in one or more non-transitory tangible media for execution and when executed operable for:

receiving, by a server device via a network, a recommendation request from a website server device having sent the recommendation request in response to a determined absence of new network content within a website service provided by the website server device, the recommendation request specifying an identified user having sent a request, via a user device and the network, to the website service, the request requesting network content that is provided within the website service;

generating by the server device an ordered list of network items most likely to be preferred by the identified user based on the network content requested by the user; and

sending the ordered list by the server device to the website server device for presentation within the website service of the ordered list to the identified user, wherein the ordered list is generated based on;

identifying user selection preferences of the identified user based on an accumulation of user selection inputs executed by the identified user relative to input options presented to the user and identifying respective available network items, the user selection inputs including selections of a first portion of the network content within the website service and having been consumed by the user,

determining first network users having a highest correlation of shared interests with the identified user based on identifying preferred network items from the available network items and having highest relative item affinity values generated for the identified user, and determining the first network users as providing highest relative user affinity values for the preferred network items,

identifying, as personally interesting content for the identified user, the preferred network items for each of the first network users based on the respective user selection preferences, and

filtering the personally interesting content for the identified user relative to socially related content, the socially related content determined based on identifying second network users having the highest relative user affinity values toward the network content specified in the request, and identifying second preferred network items for each of the second network users based on the respective user selection preferences;

the filtering of the personally interesting content for the identified user relative to socially related content based on generating an ordered list of content related to the website service, the ordered list of content generated based on an ordered sorting of the socially related content with contextually related content, the contextually related content specified in the recommendation request and identifying a second portion of the network content within the website service that has not been consumed by the user, wherein the ordered list of content is ordered based on a determined convergence of relationships between identified network items in the socially related content and the contextually related content;

wherein the determined convergence of relationships is identified based on;

identifying third network users having the highest relative user affinity values toward the second portion of the network content, and identifying third preferred network items for each of the third network users based on the respective user selection preferences, and

identifying an intersection of intersecting network items between the third preferred network items and the socially related content, and determining a social correlation between the intersecting network items and at least one of the contextually related content.