Customer self service subsystem for classifying user contexts

• US 6,853,998 B2
• Filed: 02/07/2001
• Issued: 02/08/2005
• Est. Priority Date: 02/07/2001
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Keller, Neal M., Rolando, Stephen J., Biebesheimer, Debra L., Oblinger, Daniel A., Podlaseck, Mark E.
• Primary Examiner(s): Alam, Shahid
• Assistant Examiner(s): Woo, Isaac M.
• Application Number:
  US09/778,378
• Publication Number:
  US 20020107843A1
• Time in Patent Office: 1,462 Days
• Field of Search: 707/1, 707/2, 707/3, 707/4, 707/5, 707/6, 707/7, 707/8, 707/9, 707/10, 707/100, 707/101, 707/102, 707/104.1, 707/200, 707/205
• US Class Current: 707/765
• CPC Class Codes: G06F 16/24575 using context
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access
• CPC Class Codes: Y10S 707/99936 Pattern matching access
• 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 classifying user context in a customer self service system that performs resource search and selection and 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 classifying system comprises a computing device for receiving a user query and a context vector comprising data associating an interaction state with the user and, processing the query and context vector against data included in the context attribute database and context attribute function database for predicting a particular user context. The classifier populates the user context vector with context parameters specifying a user interaction state for use in a subsequent resource search. The result of this invention is an ability to relieve the user of the nonproductive work of describing their context and the ability to improve the search value by including criteria derived from both data and behaviors in the general population which may be unknown to the user. The system and method is especially applicable for a self service system in a variety of customer self service domains including education, real estate and travel.

Claims

1. A user context classifier for a customer self service system that performs resource search and selection, said 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 classifier comprising a processing mechanism for receiving a user query and a context vector comprising data associating an interaction state with said user and including context that is a function of the user, and processing said query and context vector against data included in said context attribute database and context attribute function database for predicting a particular user context, wherein said classifier implements a supervised learning algorithm to predict user contexts and populates said user context vector with context parameters specifying a user interaction state for use in a subsequent resource search.

2. The user context classifier for a customer self service system as claimed in claim 1, wherein said processing mechanism includes mechanism for applying said functions to context for specifying said user interaction state including context that is a function of the user, said mechanism further annotating the user context vector with a set of context parameters for use in subsequent processing.

3. The user context classifier for a customer self service system as claimed in claim 1, further including mechanism for updating the context attribute value functions database with more enhanced functions.

4. The user context classifier for a customer self service system as claimed in claim 3, wherein said system further includes a user interaction database comprising data relating to past user queries entered into the system and associated user contexts for particular users, said mechanism for updating the attribute value functions database comprising mechanism for analyzing historical user interaction data from the user interaction database and learning how user context attribute values map to user context attribute functions, wherein said data from the user records database serves as a training set for supervised learning to enable continuous improvement of said functions in said context attribute function database.

5. The user context classifier for a customer self service system as claimed in claim 4, wherein said user interaction data includes data relating to previous system interactions, said data including user validated user contexts that were applicable during said prior system interactions, and the users responses relating to those interactions.

6. The user context classifier for a customer self service system as claimed in claim 4, wherein said updating mechanism provides additions and modifications to a set of context attribute functions resulting in increasing ability to predict derived contexts as functions of the raw contexts.

7. The user context classifier for a customer self service system as claimed in claim 5, wherein said previous system interaction data further includes prior transactions of a current user and prior transactions of other similar users, wherein common behaviors and acceptance criteria are determined for said updating said functions.

8. The user context classifier for a customer self service system as claimed in claim 7, wherein similar users comprise those users with shared organization, community or environmental characteristics.

9. The user context classifier for a customer self service system as claimed in claim 1, wherein said processing mechanism for receiving a user query and a context vector comprising data associating an interaction state with said user and including context that is a function of the user further includes user context including criteria beyond content.

10. The user context classifier for a customer self service system as claimed in 1, wherein said classifier further populates said user context vector with context parameters beyond content, for use in a subsequent resource search.

11. A method for classifying user contexts for a customer self service system that performs resource search and selection, said method comprising the steps of:

a) receiving a user query and a context vector comprising data associating an interaction state with said user and including context that is a function of the user;

b) processing said query and context vector against data included in 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

c) processing said query and context vector against data included in a context attribute function database comprising functions for computing values for each context attribute, wherein said processing steps b) and c) includes implementing a supervised learning algorithm for predicting a particular user context and populating said user context vector with context parameters specifying a user interaction state for use in a subsequent resource search.

12. The method as claimed in claim 11, wherein said processing step c) further includes the step of applying said functions to context for specifying said user interaction state including context that is a function of the user, said populating step including annotating the user context vector with a set of context parameters for use in subsequent processing.

13. The method as claimed in claim 11, further including the step of updating the context attribute value functions database with more enhanced functions.

14. The method as claimed in claim 13, wherein said updating step includes the steps of:

analyzing historical user interaction data from a user interaction database comprising data relating to past user queries entered into the system and associated user contexts for particular users; and

,

mapping user context attribute values to user context attribute functions, said data from said user records database serving as a training set for supervised learning to enable continuous improvement of said functions in said context attribute function database.

15. The method as claimed in claim 14, wherein said user interaction data further includes data relating to previous system interactions, said data including user validated user contexts that were applicable during said prior system interactions, and the users responses relating to those interactions.

16. The method as claimed in claim 15, wherein said previous system interactions includes prior transactions of a current user and prior transactions of other similar users, said functions updating step including the step of determining common behaviors and acceptance criteria from said previous system interactions.

17. The method as claimed in claim 16, wherein said similar users comprise those users with shared organization, community or environmental characteristics.

18. The method as claimed in claim 16, wherein said updating step includes the steps of providing additions and modifications to a set of contest attribute functions resulting in increasing ability to predict derived contexts as functions of raw contexts.

19. The method as claimed in claim 11, wherein said step a) of receiving a user query and a context vector comprising data associating an interaction state with said user and including context that is a function of the user further includes user context including criteria beyond content.

20. The method as claimed in claim 11, wherein said step c) of populating said user context vector with context parameters includes populating said user context vector with context parameters specifying a user interaction state, including parameters beyond content, for use in a subsequent resource search.

21. A program storage device readable by machine, tangibly embodying a program of instructions executable by the machine to perform method steps for classifying user contexts for a customer self service system that performs resource search and selection, said method comprising the steps of;

a) receiving a user query and a context vector comprising data associating an interaction state with said user and including context that is a function of the user;

b) processing said query and context vector against data included in 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

c) processing said query and context vector against data included in a context attribute function database comprising functions for computing values for each context attribute, wherein said processing steps b) and c) includes implementing a supervised learning algorithm for predicting a particular user context and populating said user context vector with context parameters specifying a user interaction state for use in a subsequent resource search.

22. The program storage device readable by machine as claimed in claim 21, wherein said processing step c) further includes the step of applying said functions to context for specifying said user interaction state including context that is a function of the user, said populating step including annotating the user context vector with a set of context parameters for use in subsequent processing.

23. The program storage device readable by machine as claimed in claim 21, further including the step of updating the context attribute value functions database with more enhanced functions.

24. The program storage device readable by machine as claimed in claim 23, wherein said updating step includes the steps of:

analyzing historical user interaction data from a user interaction database comprising data relating to past user queries entered into the system and associated user contexts for particular users; and

,

mapping user context attribute values to user context attribute functions, said data from said user records database serving as a training set for supervised learning to enable continuous improvement of said functions in said context attribute function database.

25. The program storage device readable by machine as claimed in claim 24, wherein said user interaction data further includes data relating to previous system interactions, said data including user validated user contexts that were applicable during said prior system interactions, and the users responses relating to those interactions.

26. The program storage device readable by machine as claimed in claim 21, wherein said step a) of receiving a user query and a context vector comprising data associating an interaction state with said user and including context that is a function of the user further includes user context including criteria beyond content.

27. The program storage device readable by machine as claimed in claim 21, wherein said step c) of populating said user context vector with context parameters includes populating said user context vector with context parameters specifying a user interaction state, including parameters beyond content, for use in a subsequent resource search.