Customer self service subsystem for context cluster discovery and validation

• US 6,873,990 B2
• Filed: 02/07/2001
• Issued: 03/29/2005
• Est. Priority Date: 02/07/2001
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Oblinger, Daniel A.
• Primary Examiner(s): Alam, Shahid
• Assistant Examiner(s): WOO, ISAAC M
• Application Number:
  US09/778,149
• Publication Number:
  US 20020107852A1
• Time in Patent Office: 1,511 Days
• Field of Search: 707/1, 707/2, 707/3, 707/4, 707/5, 707/6, 707/10, 707/101, 707/102, 707/103
• US Class Current: 1/1
• CPC Class Codes: G06F 16/337 Profile generation, learning or modification
• 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

Abstract

A system and method for clustering user contexts in a customer self service system that performs resource search and selection. The customer self service system includes a context attribute database comprising types of user contexts and one or more context attributes associated with each user context for processing by the system, and context attribute function database comprising functions for computing values for each context attribute. The context clustering system receives user interaction data from among a database of user interaction records and a distance metric for associating closeness of the user interaction data and clusters the user interaction data according to the distance metric to determine new user contexts and associated attributes for use in subsequent resource searches initiated by users in the system. Improved user query definition and resource lookup results from the new determined user context attributes. The user interaction data comprises past and present user queries, system responses to user queries, raw context information including: one or more of static, historical context, transient context, organizational context, community context, and environment context, and, other raw context associated with the user and dependent upon that user'"'"'s interaction state and customer query domain.

Claims

1. A user context clustering system for a customer self service system that performs resource search and selection, said system comprising:

a context attribute database comprising types of user contexts and one or more context attributes associated with each user context for processing by said system;

a context attribute function database comprising funtions for computing values for each context attribute;

a database of user interaction records including user interaction data; and

,

a processing mechanism for receiving said user interaction data including traces of previous user interactions with the system and including context information that is a function of said user including that user'"'"'s interaction state, and receiving a distance metric for associating closeness of said user interaction data and, for performing unsupervised learning by clustering related aspects of multiple user interactions with said system from said user interaction data according to said distance metric to determine new user contexts and associated attributes for use in subsequent resource searches initiated by users in said system, wherein improved query definition and resource lookup results from said new determined user contrxt attributes due to user validation based on a combination of said unsupervised learning and supervised learning.

2. The user context clustering system for a customer self service system as claimed in claim 1, further comprising system administrator for updating said context attribute database with determined new user contexts and associated context attributes.

3. The user context clustering system for a customer self service system as claimed in claim 1, wherein said system administrator develops new context attributes and context attribute functions for computing values for said context attributes.

4. The user context clustering system for a customer self service system as claimed in claim 1, wherein said system generates new user interaction records with said determined new context attributes.

5. The user context clustering system for a customer self service system as claimed in claim 4, wherein said system administrator generates new context attribute names from said clustering, and assigns new records in said user interaction records database with values for those attributes, said updating context assignments serve as the training data for continuously improving said functions in said context attribute functions database.

6. The user context clustering system for a customer self service system as claimed in claim 1, wherein said distance metric includes determining closeness of parameters of said user interaction data, a closeness parameter including similarity of result sets of a user query.

7. The user context clustering system for a customer self service system as claimed in claim 1, wherein said system administrator develops new definitions and logic for mapping specific resources to specific context sets, and updating a database for storing said definitions and mapping logic.

8. The user context clustering system for a customer self service system as claimed in claim 1, wherein said user interaction data comprises past and present user queries.

9. The user context clustering system for a customer self service system as claimed in claim 1, wherein said user interaction data comprises system responses to said user queries.

10. The user context clustering system for a customer self service system as claimed in claim 1, wherein said user interaction data comprises raw context information including:

one or more of static, historical context, transient context, organizational context, community context, and environment context.

11. The user context clustering system for a customer self service system as claimed in claim 10, wherein said user interaction data comprises other raw context associated with the user and dependent upon that user'"'"'s interaction state and query domain.

12. The user context clustering system for a customer self service system as claimed in claim 11, wherein a query domain includes one of:

education, travel and real estate.

13. The user context clustering system for a customer self service system as claimed in claim 1, wherein said user interaction data comprises user feedback.

14. The user context clustering system for a customer self service system as claimed in claim 1, wherein said processing mechanism for performing unsupervised learning implements an unsupervised clustering algorithm for clustering said user interaction data records.

15. A method for clustering user contexts in a customer self service system that performs resource search and selection, said customer self service system including a context attribute database comprising types of user contexts and one or more context attributes associated with each user context for processing by said system, and context attribute function database comprising functions for computing values for each context attribute, the method comprising the steps of:

a) receiving user interaction data including traces of previous user interactions with the system including context information that is a function of said user including that user'"'"'s interaction state from among a database of user interaction records and a distance metric for associating closeness of said user interaction data; and

,

b) performing unsupervised learning by clustering related aspects of multiple user interactions with said system from said user interaction data according to said distance metric to determine new user contexts and associated attributes for use in subsequent resource searches initiated by users in said system, wherein improved query definition and resource lookup result from said unsupervised learning and supervised learning.

16. The method for clustering user contexts as claimed in claim 15, further comprising the step of:

updating said context attribute database with determined new user contexts and associated context attributes.

17. The method for clustering user contexts as claimed in claim 15, further comprising the step of implementing a system administrator for developing new context attributes and context attribute functions for computing values for said context attributes.

18. The method for clustering user contexts as claimed in claim 17, further comprising the step of generating new user interaction records with said determined new context attributes.

19. The method for clustering user contexts as claimed in claim 17, wherein said clustering step b) includes the step of determining closeness of parameters of said user interaction data, a closeness parameter including similarity of result sets of a user query.

20. The method for clustering user contexts as claimed in claim 17, further comprising the step of developing new definitions and logic for mapping specific resources to specific context sets, and updating a database for storing said definitions and mapping logic.

21. The method for clustering user contexts as claimed in claim 17, further including the steps of:

generating new context attribute names from said clustering; and

,

assigning new records in said user interaction records database with values for those attributes, said updating context assignments serve as the training data for continuously improving said functions in said context attribute functions database.

22. The method for clustering user contexts as claimed in claim 15, wherein said clustering step b) of performing unsupervised clustering further includes the step of implementing an unsupervised clustering algorithm for clustering said user interaction data records.

23. A program storage device readable by machine, tangibly embodying a program of instructions executable by the machine to perform method steps for clustering user contexts for a customer self service system that performs resource search and selection, said customer self service system including a context attribute database comprising types of user contexts and one or more context attributes associated with each user context for processing by said system, and context attribute function database comprising functions for computing values for each context attribute, said method comprising the steps of:

a) receiving user interaction data including traces of previous user interactions with the system including context information that is a function of said user including that user'"'"'s interaction state from among a database of user interaction records and a distance metric for associating closeness of said user interaction data; and

,

b) performing unsupervised learning by clustering related aspects of multiple user interactions with said system from said user interaction data according to said distance metric to determine new user contexts and associated attributes for use in subsequent resource searches initiated by users in said system, wherein improved query definition and resource lookup results from said new determined user context attributes due to user validation based on a combination of said unsupervised learning and supervised learning.

24. The program storage device readable by machine as claimed in claim 23, wherein said method further comprises the step of:

updating said context attribute database with determined new user contexts and associated context attributes.

25. The program storage device readable by machine as claimed in claim 23, wherein said method further comprises the step of implementing a system administrator for developing new context attributes and context attribute functions for computing values for said context attributes.

26. The program storage device readable by machine as claimed in claim 25, wherein said method further comprises the step of updating said new user interaction records with said determined new context attributes.

27. The program storage device readable by machine as claimed in claim 25, wherein said clustering step b) includes the step of determining closeness of parameters of said user interaction data, a closeness parameter including similarity of result sets of a user query.

28. The program storage device readable by machine as claimed in claim 25, wherein said method further comprises the step of developing new definitions and logic for mapping specific resources to specific context sets, and updating a database for storing said definitions and mapping logic.

29. The program storage device readable by machine as claimed in claim 25, wherein said method further includes the steps of:

generating new context attribute names from said clustering; and

,

assigning new records in said user interaction records database with values for those attributes, said updating context assignments serve as the training data for continuously improving said functions in said context attribute functions database.

30. The program storage device readable by machine as claimed in claim 23, wherein said clustering step b) of performing unsupervised learning by clustering further includes the step of implementing an unsupervised clustering algorithm for clustering said user interaction data records.