Automated classification of items using cascade searches

• US 6,697,799 B1
• Filed: 10/24/2002
• Issued: 02/24/2004
• Est. Priority Date: 09/10/1999
• Status: Expired due to Term

• Current Assignee: International Business Machines Corporation
• Original Assignee: Requisite Technology, Inc. (Huron Consulting Group Incorporated)
• Inventors: Beall, Christopher Wade, Neal, Michael Renn, Wilmsen, James Michael
• Primary Examiner(s): ALAM, SHAHID AL
• Application Number:
  US10/279,658
• Time in Patent Office: 488 Days
• Field of Search: 707/1, 707/2, 707/3, 707/4, 707/10, 707/100, 707/102, 707/103.R, 705/26, 706/12, 706/45, 709/218, 709/224, 709/245
• US Class Current: 1/1
• CPC Class Codes: G06F 16/906 Clustering; Classification
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching

Abstract

The present invention allows an item to automatically be classified using its attributes based on a classification schema and a knowledge base. The invention can include selecting a first attribute of the item, designating a first search strategy comprising the value of the first attribute applied to operate upon data records in a first database, selecting a second attribute of the item; designating a second search strategy comprising the value of the second attribute applied to operate upon a second subset of data records in a second database, forming a search hierarchy comprised of the first search strategy followed by the second search strategy, executing the search hierarchy, and selecting at least one classification for the item based on the search.

Claims

1. A method for classifying an item, the item having a plurality of attributes, each attribute having a value, the method comprising:

selecting a first attribute of the item;

designating a fist search strategy comprising the value of the first attribute applied to operate upon data records in a first database;

selecting a second attribute of the item;

designating a second search strategy comprising the value of the second attribute applied to operate upon a second subset of data records in a second database;

forming a search hierarchy comprised of the first search strategy followed by the second search strategy;

executing the search hierarchy;

selecting at least one classification for the item based on the search.

2. The method of claim 1, wherein the first database and the second database comprise the same database.

3. The method of claim 1, wherein the first attribute and the second attribute comprise the same attribute.

4. The method of claim 1, wherein the first search strategy comprises the value of the first attribute applied to a first search methodology to operate upon a first subset of data records in the first database and the second search strategy comprises the value of the second attribute applied to a second search methodology to operate upon a second subset of data records in the second database.

5. The method of claim 4, wherein data records in the first subset are identified as being in the first subset by a certain value in the field of the data records.

6. The method of claim 4, further comprising designating a third search strategy comprising the value of the first attribute applied to operate upon a third subset of data records in the second database.

7. The method of claim 1, wherein the first search strategy comprises at least one of a fuzzy logic search, an exact match, a soundex search, a stem search, and a word count search.

8. The method of claim 1, wherein the first and second search strategies each comprise different search methodologies.

9. The method of claim 1, wherein the first database comprises an engineered knowledge base of classifications, each classification associated with at least one attribute and each attribute associated with at least one attribute value.

10. The method of claim 9, wherein the engineered knowledge base contains a plurality of item classifications mapped to a second classification schema.

11. The method of claim 1, wherein the first attribute is a part number of the item.

12. The method of claim 1, wherein the first attribute is a description of the item.

13. The method of claim 1, further comprising analyzing the first attribute value of the item against a stop list and excluding any stop list words from the comparison.

14. The method of claim 1, further comprising presenting the selected classifications to a user, receiving a designation from the user of at least one selected classification;

and classifying the item in the designated classifications.

15. The method of claim 1, further comprising determining a confidence score for each selected classification for the item.

16. The method of claim 14, further comprising comparing confidence scores for all selected classifications and classifying the item in at least one of the selected classifications based on the confidence score comparison.

17. The method of claim 14, further comprising automatically classifying the item if the confidence score is above a threshold.

18. The method of claim 1, wherein executing the search hierarchy comprises completing execution of the first search strategy before beginning execution of the second search strategy.

19. The method of claim 18, further comprising skipping execution of the second search strategy upon finding at least one match using the first search strategy.

20. The method of claim 18, further comprising terminating execution of the fist search strategy before beginning execution of the second search strategy upon finding at least one match.

21. A machine-readable medium having stored thereon data representing sequences of instructions which, when executed by a machine, cause the machine to perform operations comprising:

selecting a first attribute of an item to be classified, the item having a plurality of attributes, each attribute having a value;

designating a first search strategy comprising the value of the first attribute applied to operate upon data records in a first database;

selecting a second attribute of the item to be classified;

designating a second search strategy comprising the value of the second attribute applied to operate upon a second subset of data records in a second database;

forming a search hierarchy comprised of the first search strategy followed by the second search strategy;

executing the search hierarchy;

selecting at least one item classification for the item based on the executing the search.

22. The medium of claim 21, wherein the first database and the second database comprise the same database.

23. The medium of claim 21, wherein the first attribute and the second attribute comprise the same attribute.

24. The medium of claim 21, wherein the first search strategy comprises the value of the first attribute applied to a first search methodology to operate upon a first subset of data records in the first database and the second search strategy comprises the value of the second attribute applied to a second search methodology to operate upon a second subset of data records in the second database.

25. The medium of claim 21, wherein the first search strategy comprises at least one of a fuzzy logic search, an exact match, a soundex search, a stem search, and a word count search.

26. The medium of claim 21, wherein the first and second search strategies each comprise different search methodologies.

27. The medium of claim 21, wherein the first database comprises an engineered knowledge base of classifications, each classification associated with at least one attribute and each attribute associated with at least one attribute value.

28. The medium of claim 21, further comprising instructions which, when executed by the machine, cause the machine to perform further operations comprising determining a confidence score for each selected item classification for the item.

29. The medium of claim 28, further comprising instructions which, when executed by the machine, cause the machine to perform further operations comprising comparing confidence scores for all selected classifications and classifying the item in at least one of the selected classifications based on the confidence score comparison.

30. The medium of claim 28, further comprising instructions which, when executed by the machine, cause the machine to perform further operations comprising automatically classifying the item if the confidence score is above a threshold.

31. An apparatus for classifying an item, the item having a plurality of attributes, each attribute having a value, the apparatus comprising:

a first classification knowledge database containing a plurality of values, each associated with at least one category;

a search engine to select a first attribute and a second attribute of the item, to execute a search hierarchy including a first search strategy comprising the value of the first attribute applied to operate upon a first subset of data records in the first classification knowledge database, the search hierarchy further including a second search strategy comprising the value of the second attribute applied to operate upon a second subset of data records in the first classification knowledge database, the search engine to select at least one classification for the item based on the search.

32. The apparatus of claim 31, further comprising a second classification knowledge database.

33. The apparatus of claim 31, wherein the search hierarchy further comprises a third search strategy comprising the value of the first attribute applied to operate upon a subset of data records in the second classification knowledge database.

34. The apparatus of claim 31, wherein the first search strategy comprises at least one of a fuzzy logic search, an exact match, a soundex search, a stem search, and a word count search.

35. The apparatus of claim 31, wherein the search engine is further to determine a confidence score for each selected classification for the item.