Systems and methods for recommendation scraping

• US 10,210,559 B2
• Filed: 03/30/2015
• Issued: 02/19/2019
• Est. Priority Date: 05/17/2012
• Status: Active Grant

• Current Assignee: Walmart Apollo, LLC (WalMart Inc.)
• Original Assignee: Walmart Apollo, LLC (WalMart Inc.)
• Inventors: Stoll, Nathan, Stensmo, Jan Magnus, Robinson, Stuart Payton, Ellerhorst, Andrew Mark
• Primary Examiner(s): Shui, Ming
• Application Number:
  US14/673,413
• Publication Number:
  US 20150278916A1
• Time in Patent Office: 1,422 Days
• Field of Search: 705 261- 272
• CPC Class Codes: G06Q 30/0609 Buyer or seller confidence or verification
• CPC Class Codes: G06Q 30/0631 Item recommendations
• CPC Class Codes: G06Q 50/01 Social networking

Abstract

Various embodiments of systems and methods for recommendation scraping include a method comprising receiving a content feed, wherein the content feed is associated with a holder and identifying at least one post within the content feed comprising a mention of a respective item. In many embodiments, the method can further comprise generating an item recommendation corresponding to the respective item, associating the item recommendation with a profile associated with the holder, and sending instructions to present the item recommendation to one or more users. Other embodiments are also disclosed herein.

Claims

1. A method, comprising:

receiving, with a computer server system, a request for an item recommendation from a third user of one or more third users, wherein the computer server system comprises a recommendation system and a trust calculation system, wherein the recommendation system is configured to determine potential recommenders in an online trust network of the third user of the one or more third users, wherein the trust calculation system is configured to build a trust graph by identifying trust levels between trusted users in the online trust network of the third user based on trust data, wherein the trusted users comprise a holder, a first user, one or more second users, and the third user of the one or more third users;

receiving a computerized social media network content feed, with the computer server system, wherein the computerized social media network content feed is associated with the holder, wherein the computerized social media network content feed is configured to be scraped for product information;

identifying, with the computer server system, at least one post within the computerized social media network content feed comprising a mention of a respective item in a trust category of one or more trust categories;

determining, with the computer server system, by natural language processing or text mining, that the mention of the respective item is positive, negative, or neutral;

generating, with the computer server system, the item recommendation corresponding to the respective item based at least in part on the determining that the mention of the respective item is positive;

associating, with the computer server system, the item recommendation with a profile associated with the holder;

sending, with the computer server system, instructions to present the item recommendation to the third user of the one or more third users, the holder being within the online trust network of the third user of the one or more third users, the online trust network comprising;

trust information between the holder and one or more users, each of the one or more users comprising an associated level of trust with the holder and one or more other users of the one or more users, the one or more users comprising the one or more second users;

a first trust indication from a second user of the one or more second users to the holder, the first trust indication not being directly from the third user to the holder, the first trust indication comprises;

a first trust level from the second user of the one or more second users to the holder; and

the trust category of the one or more trust categories in which the second user trusts the holder;

a second trust indication from the third user of the one or more third users to the second user of the one or more second users, the second trust indication not being directly from the third user to the holder, the second trust indication comprising;

a second trust level from the third user to the second user; and

the trust category of the one or more trust categories in which the third user trusts the second user;

an implicit trust indication from the third user of the one or more third users to the holder based at least in part on the first trust indication from the second user of the one or more second users to the holder and the second trust indication from the third user of the one or more third users to the second user, wherein;

the third user is within a predetermined number of connections from the holder within the online trust network of the third user, and the third user is not directly connected to the holder within the online trust network of the third user;

the one or more trust categories are arranged in a hierarchy;

a level discount comprises a calculation of p(l−

r), where p comprises a trust level and r comprises a fixed amount by which the trust level is reduced, the trust level comprises the first trust level or the second trust level;

the second trust indication between the second user and the third user is propagated to a lower level of the hierarchy without the level discount;

the second trust indication between the second user and the third user is propagated to a higher level of the hierarchy with the level discount; and

the level discount further comprises a transitive trust level for the third user within a first trust graph of the first user, wherein the transitive trust level is for the third user based on;

calculating the transitive trust level for the third user in the first trust graph of the first user based at least in part on a level of trust between the first user and the second user in the first trust graph of the first user; and

discounting the transitive trust level for the third user by a predefined amount based on the level of trust between the first user and the second user;

receiving, with the computer server system, a purchase decision from the third user in response to the item recommendation from the holder; and

initiating, with the computer server system, the purchase transaction by the third user in response to the item recommendation by the holder.

2. The method of claim 1, wherein:

the profile associated with the holder corresponds to a user account associated with the holder for an online network associated with the online trust network of the third user.

3. The method of claim 1, further comprising:

determining when a user account associated with the holder for an online network associated with the online trust network of the third user is active;

when the user account associated with the holder is inactive;

generating the profile associated with the holder; and

when the user account associated with the holder is active;

corresponding the profile associated with the holder to the user account associated with the holder.

4. The method of claim 3, further comprising:

when the user account associated with the holder is inactive;

sending an invitation to the holder to activate the user account; and

corresponding the profile associated with the holder to the user account associated with the holder upon activation of the user account.

5. The method of claim 1, wherein:

the computerized social media network content feed comprises;

a weblog associated with the holder, a microblog associated with the holder, or an image collection associated with the holder.

6. The method of claim 1, wherein:

the computerized social media network content feed comprises a weblog or microblog feed associated with the holder identifying the at least one post or the mention of the respective item.

7. The method of claim 1, wherein:

the request for the item recommendation includes an indication to initiate a transaction to purchase the item recommendation.

8. The method of claim 1, further comprising:

determining if a user account associated with the holder for an online network associated with the online trust network of the third user is active;

when the user account associated with the holder is inactive;

generating the profile associated with the holder;

sending an invitation to the holder to activate the user account; and

corresponding the profile associated with the holder to the user account associated with the holder upon activation of the user account; and

when the user account associated with the holder is active;

corresponding the profile associated with the holder to the user account associated with the holder;

wherein;

the computerized social media network content feed comprises;

a weblog associated with the holder, a microblog associated with the holder, or an image collection associated with the holder; and

the request for the item recommendation includes an indication to initiate a transaction to purchase the item recommendation.

9. A method comprising:

receiving, with a computer server system, a request for an item recommendation from a third user of one or more third users, wherein the computer server system comprises a recommendation system and a trust calculation system, wherein the recommendation system is configured to determine potential recommenders in an online trust network of the third user of the one or more third users, wherein the trust calculation system is configured to build a trust graph by identifying trust levels between trusted users in the online trust network of the third user based on trust data, wherein the trusted users comprise a holder, a first user, one or more second users, and the third user of the one or more third users;

receiving a computerized social media network content feed, with the computer server system, wherein the computerized social media network content feed is associated with the holder wherein the computerized social media network content feed is configured to be scraped for product information;

identifying, with the computer server system, at least one post within the computerized social media network content feed comprising a mention of a respective item in a trust category of one or more trust categories;

determining, with the computer server system, by natural language processing or text mining, that the mention of the respective item is positive;

generating, with the computer server system, the item recommendation corresponding to the respective item based at least in part on the determining that the mention of the respective item is positive;

associating, with the computer server system, the item recommendation with a profile associated with the holder;

sending, with the computer server system, instructions to display the item recommendation to the third user of the one or more third users, the holder being within the online trust network of the third user of the one or more third users, the online trust network comprising;

trust information between the holder and one or more users, each of the one or more users comprising an associated level of trust with the holder and one or more other users of the one or more users, the one or more users comprising the one or more second users;

a first trust indication from a second user of the one or more second users to the holder, the first trust indication comprises;

a first trust level from the second user of the one or more second users to the holder; and

the trust category of the one or more trust categories in which the second user trusts the holder;

a second trust indication from the third user of the one or more third users to the second user of the one or more second users, the second trust indication comprising;

a second trust level from the third user to the second user; and

the trust category of the one or more trust categories in which the third user trusts the second user;

an implicit trust indication from the third user of the one or more third users to the holder based at least in part on the first trust indication from the second user of the one or more second users to the holder and the second trust indication from the third user of the one or more third users to the second user, wherein;

the third user is within a predetermined number of connections from the holder within the online trust network of the third user, and the third user is not directly connected to the holder within the online trust network of the third user;

the one or more trust categories are arranged in a hierarchy;

a level discount comprises a calculation of p(l−

r), where p comprises a trust level and r comprises a fixed amount by which the trust level is reduced, the trust level comprises the first trust level or the second trust level;

the second trust indication between the second user and the third user is propagated to a lower level of the hierarchy without the level discount;

the second trust indication between the second user and the third user is propagated to a higher level of the hierarchy with the level discount; and

the level discount further comprises a transitive trust level for the third user within a first trust graph of the first user, wherein the transitive trust level is for the third user based on;

calculating the transitive trust level for the third user in the first trust graph of the first user based at least in part on a level of trust between the first user and the second user in the first trust graph of the first user; and

discounting the transitive trust level for the third user by a predefined amount based on the level of trust between the first user and the second user;

receiving, with the computer server system, a purchase decision from the third user in response to the item recommendation from the holder; and

initiating, with the computer server system, the purchase transaction by the third user in response to the item recommendation by the holder.

10. The method of claim 9, wherein:

the computerized social media network content feed comprises;

a weblog associated with the holder, a microblog associated with the holder, or an image collection associated with the holder.

11. The method of claim 9, wherein:

the holder is identified as trusted by a respective user of the one or more users with respect to a first level category of the one or more trust categories; and

the first level category of the one or more trust categories comprises the respective item.

12. The method of claim 9, wherein:

the computerized social media network content feed comprises;

a weblog associated with the holder, a microblog associated with the holder, or an image collection associated with the holder;

the holder is identified as trusted by the second user of the one or more second users with respect to a first level category of the one or more trust categories; and

the first level category of the one or more trust categories comprises the respective item.

13. A system, comprising:

one or more processors; and

memory storing one or more programs to be executed by the one or more processors,

wherein the one or more programs comprise instructions for;

receiving a request for an item recommendation from a third user of one or more third users, wherein a computer server system comprises a recommendation system and a trust calculation system, wherein the recommendation system is configured to determine potential recommenders in an online trust network of the third user of the one or more third users, wherein the trust calculation system is configured to build a trust graph by identifying trust levels between trusted users in the online trust network of the third user based on trust data, wherein the trusted users comprise a holder, a first user, one or more second users, and the third user of the one or more third users;

receiving a computerized social media network content feed, wherein the computerized social media network content feed is associated with the holder, wherein the computerized social media network content feed is configured to be scraped for product information;

identifying at least one post within the computerized social media network content feed comprising a mention of a respective item in a trust category of one or more trust categories;

determining, with the computer server system, by natural language processing or text mining, that the mention of the respective item is positive;

generating the item recommendation corresponding to the respective item based at least in part on the determining that the mention of the respective item is positive;

associating the item recommendation with a profile associated with the holder;

sending instructions to present the item recommendation to the third user of the one or more third users, the holder being within the online trust network of the third user of the one or more third users, the online trust network comprising;

trust information between the holder and one or more users, each of the one or more users comprising an associated level of trust with the holder and one or more other users of the one or more users, the one or more users comprising the one or more second users;

a first trust indication from a second user of the one or more second users to the holder, the first trust indication not being directly from the third user to the holder, the first trust indication comprises;

a first trust level from the second user of the one or more second users to the holder; and

the trust category of the one or more trust categories in which the second user trusts the holder;

a second trust indication from the third user of the one or more third users to the second user of the one or more second users, the second trust indication comprising;

a second trust level from the third user to the second user; and

the trust category of the one or more trust categories in which the third user trusts the second user;

an implicit trust indication from the third user of the one or more third users to the holder based at least in part on the first trust indication from the second user of the one or more second users to the holder and the second trust indication from the third user of the one or more third users to the second user, wherein;

the third user is within a predetermined number of connections from the holder within the online trust network of the third user, and the third user is not directly connected to the holder within the online trust network of the third user;

the one or more trust categories are arranged in a hierarchy;

a level discount comprises a calculation of p(l−

r), where p comprises a trust level and r comprises a fixed amount by which the trust level is reduced, the trust level comprises the first trust level or the second trust level;

the second trust indication between the second user and the third user is propagated to a lower level of the hierarchy without the level discount;

the second trust indication between the second user and the third user is propagated to a higher level of the hierarchy with the level discount; and

the level discount further comprises a transitive trust level for the third user within a first trust graph of the first user, wherein the transitive trust level is for the third user based on;

 

calculating the transitive trust level for the third user in the first trust graph of the first user based at least in part on a level of trust between the first user and the second user in the first trust graph of the first user; and

 

discounting the transitive trust level for the third user by a predefined amount based on the level of trust between the first user and the second user;

receiving a purchase decision from the third user in response to the item recommendation from the holder; and

initiating the purchase transaction by the third user in response to the item recommendation by the holder.

14. The system of claim 13, wherein:

the profile associated with the holder corresponds to a user account associated with the holder for an online network associated with the online trust network of the third user.

15. The system of claim 13, wherein:

the one or more programs further comprise instructions for;

determining when a user account associated with the holder for an online network associated with the online trust network of the third user is active;

when the user account associated with the holder is inactive;

generate the profile associated with the holder; and

when the user account associated with the holder is active;

corresponding the profile associated with the holder to the user account associated with the holder.

16. The system of claim 15, wherein:

when the user account associated with the holder is inactive;

send an invitation to the holder to activate the user account; and

correspond the profile associated with the holder to the user account associated with the holder upon activation of the user account.

17. The system of claim 13, wherein:

the computerized social media network content feed comprises;

a weblog associated with the holder, a microblog associated with the holder, or an image collection associated with the holder.

18. The system of claim 13, wherein:

the request for the item recommendation includes an indication to initiate a transaction to purchase the item recommendation.

19. The system of claim 13, wherein:

the one or more programs further comprise instructions for;

determining when a user account associated with the holder for an online network associated with the online trust network of the third user is active;

when the user account associated with the holder is inactive;

generate the profile associated with the holder;

send an invitation to the holder to activate the user account; and

correspond the profile associated with the holder to the user account associated with the holder upon activation of the user account; and

when the user account associated with the holder is active;

correspond the profile associated with the holder to the user account associated with the holder;

the computerized social media network content feed comprises;

a weblog associated with the holder, a microblog associated with the holder, or an image collection associated with the holder; and

the request for the item recommendation includes an indication to initiate a transaction to purchase the item recommendation.