DATA CLASSIFIER SYSTEM, DATA CLASSIFIER METHOD AND DATA CLASSIFIER PROGRAM

• US 20110153615A1
• Filed: 07/29/2009
• Published: 06/23/2011
• Est. Priority Date: 07/30/2008
• Status: Active Grant

• Current Assignee: NEC Corporation
• Original Assignee: NEC Corporation
• Inventors: Tateishi, Kenji, Hosomi, Itaru, Mizuguchi, Hironori, Kusui, Dai
• Granted Patent: US 9,342,589 B2
• Time in Patent Office: Days
• US Class Current: 707/740
• CPC Class Codes: G06F 16/35 Clustering; Classification
• CPC Class Codes: G06F 16/353 into predefined classes

Abstract

A data classifier system of the present invention selects a plurality of classifications correlated to data groups so as to output classification axes based on hierarchical classifications and data groups. The data classifier system includes a basic category accumulation means, a classification axis candidate reduction means and a priority calculation means. The basic category accumulation means accumulates classifications serving as basic categories used for desired classifications in advance. The classification axis candidate reduction means selects a plurality of classifications from among classifications descendant from each basic category so as to create classification axis candidates, thus reducing classification axis candidates subjected to calculations based on data quantity of classifications and hierarchical distances of classifications. The priority calculation means calculates priority in displaying classification axis candidates, the number of which is reduced by the classification axis candidate reduction means.

Claims

1. A data classifier system which selects a plurality of classifications correlated to data groups based on hierarchical classifications and data groups so as to output classification axes, said data classifier system comprising:

a basic category accumulation means which accumulates classifications serving as basic categories used for selecting desired classifications in advance;

a classification axis candidate reduction means which selects a plurality of classifications from among classifications descendant from each basic category so as to create classification axis candidates and which reduces the number of classification axis candidates subjected to calculations based on data quantity of classifications and hierarchical distances of classifications; and

a priority calculation means which calculates priority in displaying classification i axis candidates reduced by the classification axis candidate reduction means.

2. The data classifier system according to claim 1, wherein the classification axis candidate reduction means selects classification axes satisfying either a condition in which data quantity of classifications is a predetermined value or more or a condition in which data quantity of classifications is ranked in a higher place of ordering than other classifications as well as a condition in which lengths leading to a common ancestor among classifications fall within a specific range.

3. The data classifier system according to claim 1, wherein the priority calculation means determines priority based on lengths leading to a common ancestor among classifications in a classified hierarchy.

4. The data classifier system according to claim 1, wherein the priority calculation means calculates hierarchical distances of classifications in a classified hierarchy and determines priority on classification axis candidates based on at least one of depths of classifications in the classified hierarchy, data quantity of classifications and data redundancy of classifications.

5. The data classifier system according to claim 1 further comprising a display control means which inputs classification axis candidates reduced by the classification axis candidate reduction means and priority calculated by the priority calculation means so as to perform display control on data groups, wherein the display control means displays classification axis candidates in an order of priorities, changes displayed classification axes in response to classification axis candidates which are selected, and selects or reduces data groups in response to classifications selected for each classification axis.

6. The data classifier system according to claim 1 further comprising a data retrieval means which retrieves data groups based on retrieval keywords so as to output retrieval-resultant data groups to the classification axis candidate reduction means, wherein the classification axis candidate reduction means reduces the number of classification axis candidates based on the retrieval result of the data retrieval means, so that the priority calculation means calculates priority on classification axes correlated to retrieval-resultant data groups of the data retrieval means.

7. The data classifier system according to claim 1 further comprising a data correlation means which inputs hierarchical classifications and data groups so as to correlate input classifications and data groups.

8. A data classifier system which selects a plurality of classifications correlated to data groups so as to create classification axes based on hierarchical classifications and data groups and which outputs combinations of classification axes as multidimensional classification axes, said data classifier system comprising:

a basic category accumulation means which accumulates classifications serving as basic categories used for selecting desired classifications in advance;

a classification axis candidate reduction means which creates classification axes based on combinations of classifications, each correlated to at least one data, among classifications descendant from each basic category and which reduces the number of classification axis candidates subjected to calculations based on data quantity of classifications and hierarchical distances of classifications;

a multidimensional classification axis candidate creation means which creates multidimensional classification axis candidates combining classification axis candidates reduced by the classification axis candidate reduction means; and

a multidimensional priority calculation means which calculates priority on multidimensional classification axis candidates based on hierarchical distances of classifications in a classified hierarchy.

9. The data classifier system according to claim 8, wherein the classification axis candidate reduction means selects multidimensional classification axis candidates including classifications satisfying either a condition in which data quantity of classifications on classification axes ascribed to each dimension is a predetermined value or more or a condition in which data quantity of classifications is ranked in a higher place of ordering than other classifications as well as a condition in which hierarchical distances of classifications on classification axes ascribed to each dimension, i.e. lengths leading to a common ancestor among classifications, fall within a specific range.

10. The data classifier system according to claim 8, wherein the multidimensional priority calculation means changes priority on multidimensional classification axis candidates in response to lengths leading to a common ancestor among classifications on classification axes ascribed to each dimension in the classified hierarchy.

11. The data classifier system according to claim 8, wherein the multidimensional priority calculation means calculates hierarchical distances of classifications on classification axes ascribed to each dimension in the classified hierarchy and determines priority on multidimensional classification axis candidates based on at least one of depths of classifications on classification axes ascribed to each dimension in the classified hierarchy, data quantity of classifications and data redundancy of classifications.

12. The data classifier system according to claim 8 further comprising a multidimensional display control means which inputs multidimensional classification axis candidates reduced by the classification axis candidate reduction means and priority calculated by the multidimensional priority calculation means and which performs display control on data groups in a list form or in a table form, wherein the multidimensional display control means selects multidimensional classification axis candidates so as to dispose and display classifications ascribed to each dimension in the list form or in the table form, thus displaying at least one of data quantity, data names, data attributes and characteristic words corresponding to one or plural classifications which are selected.

13. The data classifier system according to claim 8 further comprising a data retrieval means which retrieves data groups based on retrieval keywords so as to output retrieval-resultant data groups to the multidimensional classification axis candidate reduction means.

14. The data classifier system according to claim 8 further comprising a data correlation means which inputs hierarchical classifications and data groups so as to correlate input classifications and data groups.

15. A data classifier method which selects a plurality of classifications correlated to data groups so as to output classification axes based on hierarchical classifications and data groups, said data classifier method comprising:

a classification axis candidate reduction process which accumulates classifications serving as basic categories used for selecting desired classifications in a database in advance, which selects a plurality of classifications from among classifications descendant from each basic category so as to create classification axis candidates, and which reduces the number of classification axis candidates subjected to calculations based on data quantity of classifications and hierarchical distances of classifications; and

a priority calculation process which calculates priority in displaying the reduced number of classification axis candidates.

16. The data classifier method according to claim 15, wherein the classification axis candidate reduction process selects classification axes satisfying either a condition in which data quantity of classifications is a predetermined value or more or a condition in which data quantity of classifications is ranked in a higher place of ordering than other classifications as well as a condition in which lengths leading to a common ancestor among classification fall within a specific range.

17. The data classifier method according to claim 15, wherein the priority calculation process determines priority based on lengths leading to a common ancestor among classifications in the classified hierarchy.

18. The data classifier method according to claim 15, wherein the priority calculation process calculates hierarchical distances of classifications in the classified hierarchy and determines priority on classification axis candidates based on at least one of depths of classifications in the classified hierarchy, data quantity of classifications and data redundancy of classifications.

19. A data classifier method which selects a plurality of classifications correlated to data groups based on hierarchical classifications and data groups so as to create classification axes and which outputs multidimensional classification axes combining a plurality of classification axes, said data classifier method comprising:

a classification axis candidate reduction process which accumulates classifications serving as basic categories used for selecting desired classifications in a database in advance, which creates classification axes based on combinations of classifications correlated to at least one data among classifications descendant from each basic category, and which reduces the number of classification axis candidates subjected to calculations based on data quantity of classifications and hierarchical distances of classifications;

a multidimensional classification axis candidate creation process which creates multidimensional classification axis candidates each combining the reduced number of classification axis candidates; and

a multidimensional priority calculation process which calculates priority on multidimensional classification axis candidates based on hierarchical distances of classifications in a classified hierarchy.

20. The data classifier method according to claim 19, wherein the classification axis candidate reduction process selects multidimensional classification axis candidates including classifications, in which data quantity of classifications on classification axes ascribed to each dimension is predetermined value or more or in which data quantity of classification falls within a certain ratio counting from a highest value, and classifications in which hierarchical distances of classifications on classification axes ascribed to each dimension, i.e. lengths leading to a common ancestor among classifications, fall within a specific range.

21. The data classifier method according to claim 19, wherein the multidimensional priority calculation process changes priority on multidimensional classification axes in response to lengths leading to a common ancestor among classifications on classification axes ascribed to each dimension in the classified hierarchy.

22. The data classifier method according to claim 19, wherein the multidimensional priority calculation process calculates hierarchical distances of classifications on classification axes ascribed to each dimension in the classified hierarchy and determines priority on multidimensional classification axis candidates based on at least one of depths of classifications on classification axes ascribed to each dimension in the classified hierarchy, data quantity of classifications and data redundancy of classifications.

23. A data classifier program which selects a plurality of classifications correlated to data groups so as to output classification axes based on hierarchical distances and data groups, said data classifier program causing a computer having a basic category accumulation means, which accumulates classifications serving as basic categoris used for selecting desired classifications, to execute

a classification axis candidate reduction process which selects a plurality of classifications from among classifications descendant from each basic category so as to create classification axis candidates and which reduces the number of classification axis candidates subjected to calculations based on data quantity of classifications and hierarchical distances of classifications, and

a priority calculation process which calculates priority in displaying the reduced number of classification axis candidates.

24. The data classifier program according to claim 23, said data classifier program causing a computer to execute the classification axis candidate reduction process in such a way that classification axes are selected to satisfy either a condition in which data quantity of classifications is a predetermined value or more or a condition in which data quantity of classifications is ranked in a higher place of ordering than other classifications as well as a condition in which lengths leading to a common ancestor among classifications fall within a specific range.

25. The data classifier program according to claim 23, said data classifier program causing a computer to execute a display control process which inputs classification axis candidates reduced by the classification axis candidate reduction process and priority calculates by the priority calculation process so as to perform display control on data groups, wherein the display control process displays classification axis candidates in an order of priorities, changes displayed classification axes in response to selected classification axis candidates, and selects or reduces data groups in response to selected classifications on classification axes.

26. The data classifier program according to claim 23, said data classifier program causing a computer to execute a data retrieval process which retrieves data groups based on retrieval keywords so as to output retrieval-resultant data groups, wherein the classification axis candidate reduction process reduces the number of classification axis candidates based on the retrieval result of the data retrieval process, so that the priority calculation process calculates priority on classification axes correlated to data groups retrieved by the data retrieval process.

27. The data classifier program according to claim 23, said data classifier program causing a computer to execute a data correlation process which inputs hierarchical classifications and data groups so as to correlate input classifications and data groups.

28. A data classifier program which selects a plurality of classifications correlated to data groups based on hierarchical classifications and data groups so as to create classification axes and which outputs multidimensional classification axes combining a plurality of classification axes, said data classifier program causing a computer having a basic category accumulation means, which accumulates classifications serving as basic categories used for selecting desired classifications in advance, to execute

a classification axis candidate reduction process which creates classification axes based on combinations of classifications correlated to at least one data among classifications descendant from each basic category, and which reduces the number of classification axis candidates subjected to calculations based on data quantity of classifications and hierarchical distances of classifications,

a multidimensional classification axis candidate creation process which creates multidimensional classification axis candidates each combining the reduced number of classification axis candidates, and

a multidimensional priority calculation process which calculates priority on multidimensional classification axis candidates based on hierarchical distances of classifications in a classified hierarchy.

29. The data classifier program according to claim 28, said data classifier program causing a computer to execute the classification axis candidate reduction process to select multidimensional classification axes including classifications in which data quantity of classifications is a predetermined value or more or in which data quantity of classifications falls within a certain ratio counting from a highest value, as well as classifications whose lengths leading to a common ancestor fall within a specific range.

30. The data classifier program according to claim 28, said data classifier program causing a computer to execute a multidimensional display control process which inputs multidimensional classification axis candidates reduced by the classification axis candidate reduction process and priority calculated by the multidimensional priority calculation process and which performs display control on data groups in a list form or in a table form, wherein the multidimensional display control process selects multidimensional classification axis candidates so as to dispose and display classifications ascribed to each dimension in a list form or in a table form, thus displaying at least one of data quantity, data names, data attributes and characteristic words in response to one or plural classifications which are selected.

31. The data classifier program according to claim 28, said data classifier program causing a computer to execute a data retrieval process which retrieves data groups based on retrieval keywords so as to output retrieval-resultant data groups.

32. The data classifier program according to claim 28, said data classifier program causing a computer to execute a data correlation process which inputs hierarchical classifications and data groups so as to correlate input classifications and data groups.