Method and apparatus for automatic updating of user profiles

• US 7,243,105 B2
• Filed: 12/31/2002
• Issued: 07/10/2007
• Est. Priority Date: 12/31/2002
• Status: Active Grant

• Current Assignee: British Telecommunications PLC (BT Group PLC)
• Original Assignee: British Telecommunications PLC (BT Group PLC)
• Inventors: Case, Simon J, Thint, Marcus
• Primary Examiner(s): AL HASHEMI, SANA A
• Application Number:
  US10/334,117
• Publication Number:
  US 20040128301A1
• Time in Patent Office: 1,652 Days
• Field of Search: 707100-102, 707/104.1, 707/103.R, 345/327, 700/83, 700/17
• US Class Current: 1/1
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: H04L 67/306 User profiles
• CPC Class Codes: Y10S 707/99942 Manipulating data structure, e.g. compression, compaction, compilation
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface
• CPC Class Codes: Y10S 707/99944 Object-oriented database structure
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing

Abstract

Appropriate updates to a respective user profile are inferred from monitored user activity. Such updates are suitable for use with applications programs that customize services based upon data and attributes stored in a user profile. The apparatus is arranged to personalize various aspects of a preferred profile updating process on the basis of stored user preferences, applying meta-rules to apply those user preferences to the definition of a set of personalized rule weights to be applied in a fuzzy inference process that infers profile updates by applying fuzzy rules and fuzzy sets to the analysis of event statistics derived from the results of user monitoring. The personalized rule weights are applied to selectively adjust the effect of particular fuzzy rules. In addition, mapping of event statistics to fuzzy rule premises in the fuzzy inference process may be personalized through the use of user-defined fuzzy sets.

Claims

1. An apparatus for use in updating a user profile, the user profile being suitable for use in providing customized services to a respective user, the apparatus comprising:

an input for receiving event statistics relating to a user'"'"'s activity;

a first rule store for storing a first set of rules;

an inference engine for interring updates to a user profile for the user according to said first set of rules using event statistics received at said input;

a store for storing user preference data;

a store for storing a set of personalized rule weightings in respect of said user;

a second rule store for storing a second set of rules;

adjustment means for adjusting said personalized rule weightings according to said second set of rules and with reference to said stored user preference data; and

wherein said inference engine is arranged to use said personalized rule weightings in the application of said first set of rules to the inference of profile updates and thereby to determine and output a set of at least one update to the user profile.

2. An apparatus as in claim 1, wherein said stored user preference data include data providing a personalized definition of one or more parameters for input to said inference engine and wherein said inference engine further comprises means for mapping received event statistics to said personalized definitions of respective input parameters.

3. An apparatus as in claim 1 wherein said stored user preference data include data defining one or more event statistics to be calculated, said apparatus further comprising calculating means for calculating personalized event statistics with reference to said stored user preference data.

4. An apparatus as in claim 1, wherein said inference engine is a fuzzy inference engine and wherein said first set of rules comprises fuzzy rules.

5. An apparatus as in claim 1, wherein said inference engine is a fuzzy inference engine, and wherein said first set of rules comprise fuzzy rules and wherein said personalized definitions of input parameters comprise fuzzy data sets.

6. An apparatus as in claim 1 wherein said user preference data comprise one or more data elements stored with the user profile.

7. An apparatus as in claim 1 wherein the user profile comprises a plurality of terms and attributes and wherein said profile updates output by said inference engine comprise adjustments to said attributes.

8. An apparatus as in claim 6, wherein said profile updates output by said inference engine comprise adjustments to data elements of said user preference data stored within the user profile.

9. An apparatus as in claim 1 further comprising means for receiving profile updates from an external source and for aggregating said received updates with output from said inference engine.

10. A method of updating a user profile, the user profile being suitable for use in providing customized services to a respective user, the method comprising:

(i) storing a first set of rules;

(ii) generating a set of personalized rule weightings according to a second set of rules and with reference to a set of user preference data;

(iii) receiving event statistics relating to a user'"'"'s activity; and

(iv) applying an inference engine to infer and output at least one update to a profile for the user according to said first set of rules weighted according to said generated set of personalized rule weightings, using said received event statistics.