RECOMMENDING PERSONALLY INTERESTED CONTENTS BY TEXT MINING, FILTERING, AND INTERFACES

• US 20130254217A1
• Filed: 03/05/2013
• Published: 09/26/2013
• Est. Priority Date: 03/07/2012
• Status: Active Grant

• Current Assignee: UT-Battelle LLC
• Original Assignee: UT-Battelle LLC
• Inventors: Xu, Songhua
• Granted Patent: US 9,171,068 B2
• Time in Patent Office: Days
• US Class Current: 707/754
• CPC Class Codes: G06F 16/335 Filtering based on additional data, e.g. user or group profiles filtering in web context G06F16/9535, G06F16/9536
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

A personalized content recommendation system includes a client interface device configured to monitor a user'"'"'s information data stream. A collaborative filter remote from the client interface device generates automated predictions about the interests of the user. A database server stores personal behavioral profiles and user'"'"'s preferences based on a plurality of monitored past behaviors and an output of the collaborative user personal interest inference engine. A programmed personal content recommendation server filters items in an incoming information stream with the personal behavioral profile and identifies only those items of the incoming information stream that substantially matches the personal behavioral profile. The identified personally relevant content is then recommended to the user following some priority that may consider the similarity between the personal interest matches, the context of the user information consumption behaviors that may be shown by the user'"'"'s content consumption mode.

Claims

1. A method of providing personalized content recommendations using a computer comprising:

providing a user interface on a computing device that monitors a user'"'"'s information stream;

storing a personal behavioral profile in a memory of the computing device, the memory stored with user'"'"'s preferences based on a plurality of monitored past behaviors and a collaborative filtering;

filtering items in an incoming information stream with the personal behavioral profile; and

receiving only those items of the incoming information stream that match the user'"'"'s interest profile for a certain information consumption mode of the user.

2. The method of claim 1 where the computing device comprises a mobile computing device.

3. The method of claim 1 where the computing device comprises a non-mobile computing device.

4. The method of claim 1 further comprising displaying only those items of the incoming information stream that match the personal behavioral profile.

5. The method of claim 1 where the personal behavioral profile is based on a user event.

6. The method of claim 1 where the filtering comprises HTML filters that substantially removes ads, navigation bars, tags, redirection links, and navigation links from the data stored in the memory.

7. The method of claim 1 further comprising extracting a content signature based on the monitored user information stream by modeling document level user'"'"'s dwell times.

8. The method of claim 1 further comprising extracting a content signature based on the monitored user information stream by modeling dwell times.

9. The method of claim 1 further comprising extracting a content signature based on the monitored user information stream by modeling concept word level dwell times.

10. The method of claim 1 further comprising extracting a content signature based on the user'"'"'s information stream by inferring word level user dwell times.

11. The method of claim 1 further comprising identifying consumed content at a page level through a comparison of time stamps of the user'"'"'s information stored in the memory.

12. The method of claim 1 further comprising generating snapshots by appending a time stamp to a user'"'"'s behavior record, and a file parsed from the monitored user information stream.

13. The method of claim 1 where the personal behavioral profile comprises an aggregation of user profiles having similar characteristics.

14. The method of claim 1 further comprising automatically detecting a shift in the personal behavioral profile and applying a second personal behavioral profile in the filtering act.

15. A personalized content recommendation system comprising:

a client interface device configured to monitor a user'"'"'s information data stream transmitted on a computer network;

a collaborative personal interest inference engine resident to a non-transient media and configured to generate automatic predictions about the interests of the user;

a database server stored with a personal behavioral profile and user'"'"'s preferences based on a plurality of monitored past behaviors and an output of the collaborative personal interest inference engine for a plurality of user consumption modes respectively; and

a server programmed to filter items in an incoming information stream with the personal behavioral profile and programmed to identify only those items of the incoming information stream that substantially matches the personal behavioral profile.

16. The system of claim 15 where the server assigns a priority score for each selected content designated well-matched with the personal interest of a user.

17. The system of claim 15 where the client interface device resides on a mobile computing device.

18. The system of claim 15 further comprising a display for rendering items of the incoming information stream that match one of the personal behavioral profiles of a user.

19. The system of claim 15 where the personal behavioral profile is based on a user event on a computing device.

20. The system of claim 15 further comprising a plurality of HTML filters that detects and substantially removes ads, navigation bars, tags, redirection links, or navigation links from the user'"'"'s information data stream.

21. The system of claim 15 further comprising a remote computing platform programmed to extract a content signature based on the monitored user information stream by modeling document level user'"'"'s dwell times.

22. The system of claim 15 further comprising a computing platform programmed to extract a content signature based on the monitored user information stream by modeling concept word level dwell times.