Content personalization based on actions performed during a current browsing session

• US 20020010625A1
• Filed: 03/29/2001
• Published: 01/24/2002
• Est. Priority Date: 09/18/1998
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon.com, Inc.
• Inventors: Smith, Brent R., Linden, Gregory D., Zada, Nida K.
• Granted Patent: US 6,853,982 B2
• Time in Patent Office: Days
• US Class Current: 705/14
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0253 During e-commerce, i.e. online transactions
• CPC Class Codes: G06Q 30/0254 based on statistics
• CPC Class Codes: G06Q 30/0255 based on user history
• CPC Class Codes: G06Q 30/0256 User search
• CPC Class Codes: G06Q 30/0271 Personalized advertisement
• CPC Class Codes: G06Q 30/0277 Online advertisement
• CPC Class Codes: G06Q 30/0282 Rating or review of business operators or products
• CPC Class Codes: G06Q 30/0601 Electronic shopping [e-shopping]
• CPC Class Codes: G06Q 30/0603 Catalogue ordering
• CPC Class Codes: G06Q 30/0631 Item recommendations
• CPC Class Codes: G06Q 30/0633 Lists, e.g. purchase orders, compilation or processing
• CPC Class Codes: G06Q 30/0641 Shopping interfaces
• CPC Class Codes: H04L 51/02 using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface

Abstract

Various methods are disclosed for monitoring user browsing activities, and for using such information to provide session-specific item recommendations to users. In one embodiment, a monitoring component of a merchant'"'"'s Web site maintains a record of products viewed by each user during a current browsing session—preferably based on visits to product detail pages. A recommendations component uses the resulting history of viewed products to identify additional products to recommend, preferably using a pre-existing table that maps products to related products. In one embodiment, this table is generated by periodically analyzing user browsing histories to identify correlations between purchases, viewing events, and/or other actions performed with respect to particular products. The recommended items may be displayed together with an option to individually deselect the recently viewed items on which the recommendations are based. Embodiments are also disclosed that use recent browse node visits, and recent searches, to generate the recommendations.

Claims

1. A method of assisting a user in locating products while browsing an online catalog of products, comprising, during a browsing session:

maintaining a history of products viewed by the user during the browsing session (“

viewed products”

);

selecting a set of additional products to present to the user based on the history of viewed products, wherein a product is selected to include in the list of additional products based at least in part on whether that product is related to more than one of the viewed products;

outputting to the user the list of the additional products and a list of the viewed products; and

presenting to the user an option to individually de-select the viewed products and to thereafter view a refined set of additional products which reflects de-selection of one or more of the viewed products.

2. The method of claim 1, wherein the product is selected to include in the list of additional products based further upon a degree of relatedness of the product to a viewed product.

3. The method of claim 1, wherein selecting a set of additional products to present to the user comprises taking into consideration a frequency with which a candidate additional product co-occurs with a viewed product within product viewing histories of users.

4. The method of claim 1, wherein selecting a set of additional products to present to the user comprises taking into consideration a frequency with which a candidate additional product co-occurs with a viewed product with product purchase histories of users.

5. The method of claim 1, wherein maintaining a history of the products viewed by the user comprises treating an access by the user to a product detail page as a viewing event with respect to a corresponding product.

6. The method of claim 1, wherein maintaining a history of the products viewed by the user comprises treating a product as viewed only when the user accesses a detail page of the product.

7. The method of claim 1, wherein maintaining a history of the products viewed by the user comprises treating an access by the user to at least one of (a) a search results page, and (b) a browse node page, as a viewing event with respect to at least one product displayed within such page.

8. A method of recommending items to a user, comprising:

providing a table that maps each of multiple items to a corresponding set of related items;

monitoring a browsing session of a user to identify a plurality of items accessed by the user during the session, wherein an item is treated as accessed at least if the user accesses a detail page of the item;

for each of the plurality of items accessed by the user, identifying within the table a corresponding set of related items, to thereby identify a plurality of sets of related items;

combining the plurality of sets of related items to form a ranked set of related items, in which a ranking of a related item reflects whether the ranked item is included within more than one of the plurality of sets of related items;

and

recommending at least one item from the ranked set to the user during the browsing session.

9. The method of claim 8, wherein each item is a product within an electronic catalog of products sold by an online merchant.

10. The method of claim 9, wherein monitoring the browsing session comprises treating a product as accessed only if the user affirmatively requests information about the product.

11. The method of claim 8, wherein combining the plurality of sets comprises filtering out items already accessed by the user.

12. The method of claim 8, wherein recommending at least one item from the ranked set comprises outputting to the user a link to a customized page that includes representations of recommended items selected from the ranked set.

13. The method of claim 12, further comprising:

providing, within the customized page, a representation of the plurality of items accessed by the user during the session, together with an option for the user to de-select each such accessed item;

detecting de-selection of one or more of the plurality of accessed items;

and

generating and outputting to the user a refined set of recommended items that reflects said de-selection of one or more of the accessed items.

14. The method of claim 8, wherein monitoring the browsing session comprises storing, during the browsing session, an identification of each of the plurality of items accessed by the user within a session record.

15. The method of claim 14, wherein the session record is stored in a volatile random access memory of a server computer.

16. The method of claim 14, wherein the session record is stored without any identification of the user such that user privacy is preserved.

17. The method of claim 8, wherein the browsing session is a voice-based browsing session in which items are accessed by requesting and listening to information about the items.

18. The method of claim 8, wherein providing a table that maps each of multiple items to a corresponding set of related items comprises generating the table by at least analyzing session-specific product viewing histories of a plurality of users to identity items that co-occur within the product viewing histories relatively frequently.

19. The method of claim 8, wherein providing a table that maps each of multiple items to a corresponding set of related items comprises generating the table by at least analyzing user purchase histories of a plurality of users to identify items that co-occur within the purchase histories relatively frequently.

20. A method of recommending items to a user, comprising:

monitoring a browsing session of the user to identify a plurality of items accessed by the user during the session (“

accessed items”

), wherein accesses to item detail pages are counted as accesses to corresponding items;

during the browsing session, outputting a link to the user to a customized page that includes representations of a set of recommended items selected based on the plurality of accessed items, the recommended items being different from the accessed items;

generating and outputting the customized page in response to selection of the link by the user; and

providing an option for the user to de-select each of the plurality of accessed items to substantially eliminate an effect of an accessed item on the set of recommended items represented within the customized web page, whereby the user may initiate generation of a refined set of recommended items.

21. The method of claim 20, wherein each item is a product within an electronic catalog of products sold by an online merchant.

22. The method of claim 20, wherein the customized page is generated and displayed to the user regardless of whether the user has added an item to a shopping cart.

23. The method of claim 20, wherein the customized page is generated and displayed to the user regardless of whether the user is recognized.

24. The method of claim 20, wherein monitoring the browsing session comprises maintaining, during the browsing session, a session-specific record within a volatile memory of a server computer, and wherein generating and outputting the customized page comprises accessing the session-specific record to identify the plurality of accessed items.

25. The method of claim 20, further comprising selecting the recommended items to include within the customized page by at least:

for each of the plurality of items accessed by the user, identifying a corresponding set of related items, to thereby identify a plurality of sets of related items;

combining the plurality of sets of related items to form a ranked set of related items, in which a ranking of a related item reflects whether the ranked item is included within more than one of the plurality of sets of related item; and

selecting at least some of the related items from the ranked set.

26. The method of claim 20, further comprising determining whether to include a first item in the set of recommended items based on whether, and a degree to which, the first item is related to each of the plurality of accessed items.

27. A system for recommending items to a user, comprising:

a monitoring component configured to monitor a browsing session of the user and to maintain a real time session record reflecting items accessed by the user during the browsing session, wherein the monitoring component treats an access by the user to an item'"'"'s detail page as an access to the item;

a non-user-specific table that maps each of a plurality of items to a corresponding set of related items; and

a recommendations component that uses the real time session record and the table to provide personalized recommendations of items to the user during the browsing session.

28. The system of claim 27, wherein the recommendations component selects an item to recommend to a user based at least in part upon whether that item is related to more than one of the items accessed by the user during the browsing session.

29. The system of claim 27, wherein the recommendations component is configured to generate the recommendations by at least:

accessing the real time session record to identify a plurality of items accessed by the user during the browsing session;

for each of the plurality of accessed items, looking up from the table a corresponding set of related items, to thereby identify multiple sets of related items; and

selecting a related item to recommend to the user based at least in part upon whether the related item is included within more than one of the multiple sets.

30. The system of claim 27, wherein the monitoring component maintains the real time session record within a volatile memory of a server computer.

31. The system of claim 27, wherein each item is a product within an electronic catalog of products sold by an online merchant.

32. The system of claim 27, wherein the recommendations component provides an option for the user to de-select an accessed item to cause the accessed item to be substantially eliminated from consideration in generating the personalized recommendations.

33. The system of claim 27, further comprising a component that generates the table by analyzing a plurality of user-specific histories of item-affinity-evidencing activities.

34. A method of assisting a user in locating products while browsing an online catalog of products in which products are arranged by category within browse nodes of a browse tree, the method comprising, during a browsing session:

maintaining a history of browse nodes accessed by the user during the browsing session (“

accessed browse nodes”

);

selecting a set of products to present to the user, wherein a product is selected to be included in the set of products based at least in part on whether that product is a member of more than one of the accessed browse nodes; and

presenting the set of products to the user during the browsing session.

35. The method of claim 34, further comprising presenting to the user an option to individually de-select the accessed browse nodes and to thereafter view a refined set of products which reflects de-selection of one or more of the accessed browse nodes.

36. A method of assisting a user in locating products while browsing an online catalog of products, the method comprising, during a browsing session:

maintaining a history of searches performed by the user during the browsing session;

selecting a set of products to present to the user, wherein a product is selected to be included in the set of products based at least in part on whether that product is included within results of more than one of the searches; and

presenting the set of products to the user during the browsing session.

37. The method of claim 36, wherein presenting the set of products to the user comprises presenting the set of products separately from the results of the searches.

38. The method of claim 36, further comprising presenting to the user an option to individually de-select the searches and to thereafter view a refined set of products which reflects de-selection of one or more of the searches.