Methods and apparatus for user-centered class supervision

• US 6,804,669 B2
• Filed: 08/14/2001
• Issued: 10/12/2004
• Est. Priority Date: 08/14/2001
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Aggarwal, Charu C.
• Primary Examiner(s): Robinson, Greta
• Assistant Examiner(s): LEWIS, CHERYL RENEA
• Application Number:
  US09/929,181
• Publication Number:
  US 20030037025A1
• Time in Patent Office: 1,155 Days
• Field of Search: 707/1, 707/7, 707/100, 707/2, 707/104.1, 707/103-R-1, 707/3, 707/4, 707/6, 706/45, 706/46, 704/243, 704/252, 704/253
• US Class Current: 1/1
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 2216/03 Data mining
• CPC Class Codes: Y10S 707/99936 Pattern matching access

Abstract

Techniques are provided for incorporating human or user interaction in accordance with the design and/or performance of data mining applications such as similarity determination and classification. Such user-centered techniques permit the mining of interesting characteristics of data in a data or feature space. For example, such interesting characteristics that may be determined in accordance with the user-centered mining techniques of the invention may include a determination of similarity among different data objects, as well the determination of individual class labels. These techniques allow effective data mining applications to be performed in accordance with high dimensional data.

Claims

1. A computer-based method of classifying a test instance in accordance with a data set, the method comprising the steps of:

obtaining a test instance;

presenting to the user at least one projection representing a distribution of the data set;

the user isolating a portion of the data presented in the at least one projection based on a relationship between the test instance and the data presented in the at least one projection;

determining a behavior of the isolated portion of data; and

determining a class for the test instance based on the isolated portion of data when the user makes a decision to do so based on the determined behavior of the isolated portion of data.

2. The method of claim 1, wherein the presenting step further comprises the user being presented with two or more projections respectively representing distributions of the data set such that the user may select one of the projections to be used in the isolating step.

3. The method of claim 1, wherein the at least one projection presented to the user is chosen randomly.

4. The method of claim 1, further comprising the step of considering at least one of another portion of the data set and a subset of the isolated portion of the data when the user makes a decision not to have a class determined for the test instance based on the isolated portion of data.

5. The method of claim 4, wherein further consideration of the data is determined by the distribution of the isolated data set.

6. The method of claim 1, wherein the class determining step is achieved by computing the fraction of the data which satisfies a particular class.

7. Apparatus for classifying a test instance in accordance with a data set, the method comprising the steps of:

at least one processor operative to;

(i) obtain a test instance;

(ii) present to the user at least one projection representing a distribution of the data set;

(iii) permit the user to isolate a portion of the data presented in the at least one projection based on a relationship between the test instance and the data presented in the at least one projection;

(iv) determine a behavior of the isolated portion of data; and

(v) determine a class for the test instance based on the isolated portion of data when the user makes a decision to do so based on the determined behavior of the isolated portion of data; and

memory, operatively coupled to the at least one processor, for storing at least one of the data set and the determined class of the test instance.

8. The apparatus of claim 7, wherein the presenting operation further comprises the user being presented with two or more projections respectively representing distributions of the data set such that the user may select one of the projections to be used in the isolating operation.

9. The apparatus of claim 7, wherein the at least one projection presented to the user is chosen randomly.

10. The apparatus of claim 7, wherein the at least one processor is further operative to permit consideration of at least one of another portion of the data set and a subset of the isolated portion of the data when the user makes a decision not to have a class determined for the test instance based on the isolated portion of data.

11. The method of claim 10, wherein further consideration of the data is determined by the distribution of the isolated data set.

12. The apparatus of claim 7, wherein the class determining operation is achieved by computing the fraction of the data which satisfies a particular class.

13. An article of manufacture for classifying a test instance in accordance with a data set, the article comprising a machine readable medium containing one or more programs which when executed implement the steps of:

obtaining a test instance;

presenting to the user at least one projection representing a distribution of the data set;

the user isolating a portion of the data presented in the at least one projection based on a relationship between the test instance and the data presented in the at least one projection;

determining a behavior of the isolated portion of data; and

determining a class for the test instance based on the isolated portion of data when the user makes a decision to do so based on the determined behavior of the isolated portion of data.

14. The article of claim 13, wherein the presenting step further comprises the user being presented with two or more projections respectively representing distributions of the data set such that the user may select one of the projections to be used in the isolating step.

15. The article of claim 13, wherein the at least one projection presented to the user is chosen randomly.

16. The article of claim 13, further comprising the step of considering at least one of another portion of the data set and a subset of the isolated portion of the data when the user makes a decision not to have a class determined for the test instance based on the isolated portion of data.

17. The article of claim 16, wherein further consideration of the data is determined by the distribution of the isolated data set.

18. The article of claim 13, wherein the class determining step is achieved by computing the fraction of the data which satisfies a particular class.