Intelligent classification system

• US 20040083191A1
• Filed: 12/27/2002
• Published: 04/29/2004
• Est. Priority Date: 10/25/2002
• Status: Abandoned Application

• Current Assignee: SAP AG (SAP SE)
• Original Assignee: SAP AG (SAP SE)
• Inventors: Kaiser, Matthias, Ronnewinkel, Christopher
• Application Number:
  US10/330,402
• Publication Number:
  US 20040083191A1
• Time in Patent Office: Days
• US Class Current: 706/20
• 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 16/353 into predefined classes
• CPC Class Codes: G06F 16/954 Navigation, e.g. using categorised browsing
• CPC Class Codes: G06Q 10/10 Office automation; Time management

Abstract

A system is disclosed to provide intelligent classification services. The system includes a classifier that provides one or more recommendations based on an incoming message. The system may include a user application that allows an incoming message to be processed by the classifier and may be utilized to respond to incoming messages.

Claims

1. A method comprising:

receiving a message including a request for information;

classifying the request for information based upon features of the message; and

providing a recommendation based upon the classification.

2. The method of claim 1 wherein the request for information comprises a problem description and the recommendation comprises a recommended solution.

3. The method of claim 1 wherein the recommendation comprises a list of suggested entities based on the request for information.

4. The method of claim 3 wherein the recommendation further comprises one or more identifiers, each of which corresponds to a respective group of one or more individuals knowledgeable about subject matter of the request for information and the method further comprising routing the message to the one or more individuals.

5. The method of claim 1 wherein classifying the request for information comprises:

comparing the request for information with previous requests for information; and

determining which previous requests are most similar to the request for information.

6. The method of claim 5 wherein determining which previous requests are most similar to the request for information comprises using a k-nearest-neighbor algorithm.

7. The method of claim 1 wherein providing a recommendation comprises:

generating a classification result using a scored list of candidate classes as input;

calculating an accuracy measure using class-weights associated with candidate classes as input; and

comparing the accuracy measure to a predetermined value.

8. The method of claim 7 further comprising displaying a class-score indicating a text-mining similarity with the request for information.

9. The method of claim 7 further comprising displaying messages from the candidate classes.

10. The method of claim 7 further comprising sending a recommendation based on the accuracy measure and the predetermined value comparison.

11. The method of claim 1 comprising updating a repository of collected examples that stores requests for information in response to a user providing a recommendation in response to the request for information.

12. The method of claim 1 comprising associating a tag value to a class equivalent as indicia of relevance to a class-center.

13. An article comprising a machine-readable medium storing machine-executable instructions that, when executed, cause a machine to perform operations comprising:

classify a request for information included in a received message based upon features of the message; and

provide a recommendation based upon the classification of the message.

14. The article of claim 13 including instructions, that when applied to the machine, cause the machine to provide a recommended solution in response to a problem description in the received message based on the classification.

15. The article of claim 13 including instructions, that when applied to the machine, cause the machine to provide a list of suggested entities in response to the request for information based on the classification.

16. The article of claim 13 wherein the recommendation comprises one or more identifiers, each of which corresponds to a respective group of one or more individuals knowledgeable about subject matter of the request for information including instructions, that when applied to the machine, cause the machine to route the message to the one or more individuals.

17. The article of claim 13 including instructions, that when applied to the machine, cause the machine to:

compare the request for information with previous requests for information; and

determine which previous requests are most similar to the request for information.

18. The article of claim 17 including instructions, that when applied to the machine, cause the machine to use a k-nearest-neighbor algorithm to determine which previous requests are most similar to the request for information.

19. The article of claim 13 including instructions, that when applied to the machine, cause the machine to:

generate a classification result using as input a scored list of candidate classes;

calculate an accuracy measure using class-weights associated with candidate classes as input; and

compare the accuracy measure to a predetermined value.

20. The article of claim 17 including instructions, that when applied to the machine, further cause the machine to display a class-score indicating a text mining similarity with the request for information.

21. The article of claim 17 including instructions, that when applied to the machine, further cause the machine to display messages from the candidate classes.

22. The article of claim 17 including instructions, that when applied to the machine, further cause the machine to send a recommendation based on the accuracy measure and the predetermined value comparison.

23. The article of claim 13 including instructions, that when applied to the machine, cause the machine to update a repository of collected examples that stores requests for information in response to a user providing a recommendation in response to the request for information.

24. The article of claim 13 including instructions, that when applied to the machine, cause the machine to associate a tag value to a class equivalent as an indicia of relevance to a class-center.

25. A system comprising:

a computer network;

a directory coupled to the network, the directory storing a knowledge base and a repository for collected examples;

a user interface coupled to the network to allow a user independently to request and receive recommendations; and

a service delivery device coupled to the network,

wherein the service delivery device includes a processor and memory storing instructions that, in response to receiving a request for information, cause the processor to;

classify the request based upon features of the message; and

provide a recommendation based upon the classification.

26. The system of claim 25 wherein the memory stores instructions that, in response to receiving the request, causes the processor to provide a recommended solution in response to a problem description in the received message based on the classification.

27. The system of claim 25 wherein the memory stores instructions that, in response to receiving the request, cause the processor to provide a list of suggested entities in response to the request for information based on the classification.

28. The system of claim 27 wherein the recommendation comprises one or more identifiers, each of which corresponds to a respective group of one or more individuals knowledgeable about subject matter of the request for information including instructions, that when applied to the machine, cause the machine to route the message to the one or more individuals.

29. The system of claim 25 wherein the memory stores instructions that, in response to receiving the request, cause the processor to:

compare the request for information with previous requests for information; and

determine which previous requests are most similar to the request for information.

30. The system of claim 29 wherein the memory stores instructions that, in response to receiving the request, cause the processor to use a k-nearest-neighbor algorithm to determine which previous requests are most similar to the request for information.

31. The system of claim 25 wherein the memory stores instructions that, in response to receiving the request, cause the processor to:

generate a classification result using a scored list of candidate classes as input;

calculate an accuracy measure using class-weights associated with candidate classes as input; and

compare the accuracy measure to a predetermined value.

32. The system of claim 31 wherein the memory stores instructions that, in response to receiving the request, further cause the processor to display a class-score indicating a text mining similarity with the request for information.

33. The system of claim 31 wherein the memory stores instructions that, in response to receiving the request, further cause the processor to display messages from the candidate classes.

34. The system of claim 31 wherein the memory stores instructions that, in response to receiving the request, further cause the processor to send a recommendation based on the accuracy measure and the predetermined value comparison.

35. The system of claim 25 wherein the memory stores instructions that, in response to receiving the request, further cause the processor to update a repository of collected examples that stores requests for information in response to a user providing a recommendation in response to the request for information.

36. The system of claim 25 wherein the memory stores instructions that, in response to receiving the request, cause the processor to associate a tag value to a class equivalent as an indicia of relevance to a class-center.