Adaptive filtering to adjust automated selection of content using weightings based on contextual parameters of a browsing session

Patent metadata

Field	Value
Publication number	US-9699490-B1
Application number	US-201514660775-A
Country	US
Kind code	B1
Filing date	Mar 17, 2015
Priority date	Feb 11, 2015
Publication date	Jul 4, 2017
Grant date	Jul 4, 2017

Abstract

This disclosure provides examples of computer-implemented systems and processes for determining a user browsing session context, intent, or activity based on contextual information associated with a user's browsing session, and using this determined context as a basis to filter or improve selection of content items for the user. For example, if the user is building up a watch list (for example, discovering new titles), then it may be desirable to filter out items for recommendation which are already on that watch list. However, when the user is searching for something to watch in the current session (for example, consuming content), then it may be more appropriate to show items already added to the watch list as recommended content. Item recommendations can thus be improved or filtered based in part on parameters of the user's browsing activity with respect to content items.

First claim

What is claimed is:

1. A system comprising:
an electronic data store configured to at least store contextual information for browse sessions of users with respect to digital content items available from an electronic catalog; and
a computing system comprising one or more hardware computing devices, said computing system in communication with the electronic data store and configured to at least:
access, from the electronic data store, contextual information associated with a browse session for a user, wherein the contextual information includes one or more attributes indicative of a user browsing context for the browse session;
determine the user browsing context with respect to items available from the electronic catalog based at least in part on the contextual information;
generate a candidate list of candidate video content recommendation items for the user such that the candidate list includes (i) one or more items from a watch list associated with the user and (ii) one or more additional items not in the watch list, the watch list indicative of one or more items previously added to the watch list to be watched later and when the one or more items were added to the watch list;
determine, based at least in part on an amount of time elapsed between (i) when a given item other than those included in the candidate list is added to the watch list associated with the user and (ii) when the given item is watched by the user the user, a time interval by which to filter the candidate list;
remove one or more candidate items that do not satisfy the determined time interval from the candidate list; and
provide one or more video content item recommendations from the candidate list that does not include the one or more removed candidate items for presentation to the user.

2. The system of claim 1 , wherein the determined user browsing context is indicative of the user exploring video content items to add to the watch list in a content discovery mode during the browse session of the user.

3. The system of claim 2 , wherein the contextual information indicates that a user computing device used for the browse session of the user is a tablet computing device or a smart phone, wherein the computing system is further configured to determine the user browsing context indicative of the user being in the content discovery mode based at least in part on the indication that the user computing device used for the browse session of the user is a tablet computing device or a smart phone.

4. The system of claim 1 , wherein the determined user browsing context is indicative of the user watching video content items in a content consumption mode during the browse session of the user.

5. The system of claim 4 , wherein the contextual information indicates that a user computing device used for the browse session of the user is a laptop computer or a television, wherein the computing system is further configured to determine the user browsing context indicative of the user being in the content consumption mode based at least in part on the indication that the user computing device used for the browse session of the user is a laptop computer or a television.

6. A system comprising:
a computing system comprising one or more hardware computing devices, said computing system configured to at least:
access contextual information associated with a browse session for a user, wherein the contextual information includes one or more attributes indicative of a user browsing context for the browse session;
determine the user browsing context with respect to items available from an electronic catalog based at least in part on the contextual information;
generate a candidate list of candidate digital content recommendation items for the user such that the candidate list includes (i) one or more items from a watch list associated with the user and (ii) one or more additional items not in the watch list, the watch list indicative of one or more items added to the watch list and when the one or more items were added to the watch list;
determine, based at least in part on an amount of time elapsed between (i) when a given item is added to the watch list and (ii) when the given item is watched, a time interval by which to filter the candidate list;
remove one or more candidate items that do not satisfy the determined time interval from the candidate list; and
output one or more digital content item recommendations from the candidate list that does not include the one or more removed candidate items.

7. The system of claim 6 , wherein the determined user browsing context is indicative of the user exploring video content items to add to the watch list during the browse session of the user.

8. The system of claim 6 , wherein the determined user browsing context is indicative of the user consuming video content items during the browse session of the user.

9. The system of claim 6 , wherein the contextual information is indicative of a type of user computing device used for the browse session of the user.

10. The system of claim 6 , wherein the computing system is further configured to:
remove a candidate item from the candidate list, wherein the candidate item was added to the watch list at a previous time period that precedes a current time period by less than the determined time interval.

11. The system of claim 6 , wherein the computing system is further configured to:
apply a respective weighting factor to each candidate item in the candidate list, based on the determined user browsing context, the determined time interval, and a time period associated with the candidate item; and
reorder the candidate items in the candidate list based on the respective weighting factors.

12. The system of claim 6 , wherein the computing system is further configured to:
determine an average elapsed time between (1) the user adding an item to the user's watch list and (2) the user actually watching the item.

13. The system of claim 6 , wherein the computing system is further configured to:
determine an average elapsed time across a plurality of users between (1) a respective user in the plurality of users adding an item to the respective user's watch list and (2) the respective user actually watching the item.

14. The system of claim 6 , wherein the computing system is further configured to:
determine an average elapsed time across a plurality of users other than the user between (1) a respective user in the plurality of users adding a specific item in the candidate list associated with the user to the respective user's watch list and (2) the respective user actually watching the specific item.

15. Non-transitory physical computer storage comprising computer-executable instructions stored thereon that, when executed by a hardware processor, are configured to perform operations comprising:
accessing, from an electronic data store, contextual information associated with a browse session for a user, wherein the contextual information includes one or more attributes indicative of a user browsing context for the browse session;
determining the user browsing context with respect to items available from the electronic catalog based at least in part on the contextual information;
generating a candidate list of candidate video content recommendation items for the user such that the candidate list includes (i) one or more items from a watch list associated with the user and (ii) one or more additional items not in the watch list, the watch list indicative of one or more items added to the watch list and when the one or more items were added to the watch list;
determin

Assignees

• Amazon Tech Inc

Inventors

• Japp Robert Philip

Classifications

• H04N21/4668

  for recommending content, e.g. movies · CPC title

• H04N21/251 Primary

  Learning process for intelligent management, e.g. learning user preferences for recommending movies (details of learning user preferences for the retrieval of video data in a video database G06F16/739; computer systems using learning methods G06N3/08) · CPC title

• H04L67/535

  Tracking the activity of the user (network monitoring arrangements H04L43/00; recording of computer activity G06F11/34) · CPC title

• G06F16/70

  of video data · CPC title

• H04N21/44213

  Monitoring of end-user related data (arrangements for monitoring the users' behaviour or opinions in broadcast systems H04H60/33) · CPC title