Data classifier system, data classifier method and data classifier program stored on storage medium

• US 9,342,589 B2
• Filed: 07/29/2009
• Issued: 05/17/2016
• Est. Priority Date: 07/30/2008
• Status: Active Grant

• Current Assignee: NEC Corporation
• Original Assignee: NEC Corporation
• Inventors: Mizuguchi, Hironori, Tateishi, Kenji, Hosomi, Itaru, Kusui, Dai
• Primary Examiner(s): KUDDUS, DANIEL A
• Application Number:
  US13/056,030
• Publication Number:
  US 20110153615A1
• Time in Patent Office: 2,484 Days
• Field of Search: 707/740, 707/797, 707/956
• US Class Current: 1/1
• 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 refers to a classified hierarchy establishing a parent-child relationship between a plurality of classifications to produce a plurality of classification axes based on classifications and data groups input to the data classifier system, said data classifier system comprising:

hardware, including a processor and memory, wherein;

the hardware is configured to implement a basic category accumulation part which accumulates classifications serving as basic categories on which basis desired classifications are selected;

the hardware is configured to implement a classification axis candidate reduction part which selects a plurality of classifications from among classifications descendant from each basic category in consideration of semantic relevancy therebetween, to produce a plurality of classification axis candidates based on data quantity of classifications with reference to a common ancestor among classifications in the classified hierarchy; and

the hardware is configured to implement a priority calculation part which calculates priority in displaying the plurality of classification axis candidates with reference to hierarchical distances, each representing a length of a path reaching the common ancestor among classifications associated with classification axis candidates,

wherein one of;

(a) the priority concerning a basic category X related to a plurality of classifications C in the classified hierarchy is calculated by

Priority (X;

C) =1/Max(X)×

1/(2×

N)×

Σ

ComDist(ci,cj),

where Max(X) denotes a depth of a deepest classification among classifications descending from the basic category X;

N denotes a number of the plurality of classifications; and

ComDist(ci,cj) denotes a distance between classifications ci and cj selected from among the plurality of classifications; and

(b) the priority concerning a basic category X related to a plurality of classifications C in the classified hierarchy is calculated by

Priority (X;

C)=W1×

Independence (X;

C)+W2×

Specifics (X;

C) +W3×

Exhausitivity (X;

C)+W4×

Uniqueness (X;

C),

where W1, W2, W3, and W4 denote weight coefficients to indexes which are determined in advance.

2. The data classifier system according to claim 1, wherein the classification axis candidate reduction part selects the plurality of classification axis candidates corresponding to classification axes with data quantity equal to or above a predetermined value or with highest and higher ranks of data quantities, wherein the plurality of classification axis candidates is reduced within a specific range of lengths of paths reaching the common ancestor in the classified hierarchy.

3. The data classifier system according to claim 1, wherein the priority calculation part calculates lengths of paths reaching the common ancestor in the classified hierarchy and determines priority for classification axis candidates based on lengths of paths reaching the common ancestor in the classified hierarchy, data quantity of classifications, or data redundancy of classifications.

4. The data classifier system according to claim 1 further comprising a display control part which performs display control based on the priority associated with data groups with reference to the plurality of classification axis candidates, wherein the display control part displays a plurality of classification axis candidates in an order of priority to allow a user to change at least one of the plurality of classification axis candidates or to select data groups for classification axes.

5. The data classifier system according to claim 1 further comprising a data retrieval part which retrieves data groups based on retrieval keywords to output retrieved data groups to the classification axis candidate reduction part, wherein the classification axis candidate reduction part reduces the plurality of classification axis candidates based on the retrieved data groups, so that the priority calculation part calculates priority for classification axes correlated to the retrieved data groups.

6. The data classifier system according to claim 1 further comprising a data correlation part which establishes correlation between the classifications and the data groups input.

7. The data classifier system according to claim 1, wherein the classification axis candidate reduction part selects a plurality of classifications constituting each classification axis candidate while precluding any classification descendant from a same ancestor.

8. The data classifier system according to claim 1, wherein the priority calculation part calculates the priority in consideration of a depth of each classification in the classified hierarchy.

9. The data classifier system according to claim 1, wherein the priority concerning a basic category X related to a plurality of classifications C in the classified hierarchy is calculated by

Priority (X:

C)=W1×

Independence (X;

C) +W2×

Specifics (X;

C) +W3×

Exhausitivity (X;

C) +W4×

Uniqueness (X;

C),

where W1, W2, W3, and W4 denote weight coefficients to indexes which are determined in advance,

and wherein the Independent(X;

C) is given by

Independence (X;

C)=1/Max(X) ×

1/(2×

N) ×

Σ

ComDist(ci,cj),

where Max(X) denotes a depth of a deepest classification among classifications descending from the basic category X;

N denotes a number of the plurality of classifications; and

ComDist(ci,cj) denotes a distance between classifications ci and cj selected from among the plurality of classifications.

10. The data classifier system according to claim 1, wherein the priority concerning a basic category X related to a plurality of classifications C in the classified hierarchy is calculated by

Priority (X:

C)=W1×

Independence (X;

C) +W2×

Specifics (X;

C) +W3×

Exhausitivity (X;

C) +W4×

Uniqueness (X;

C),

where W1, W2, W3, and W4 denote weight coefficients to indexes which are determined in advance,

and wherein the Specifics(X;

C) is given by

Specifics (X;

C)=1/Max(X)×

1/N 1/×

Σ

Depth(X,ci),

where Max(X) denotes a depth of a deepest classification among classifications descending from the basic category;

N denotes a number of the plurality of classifications; and

Depth(X,ci) denotes a shortest path length from the basic category X to a classification ci selected from among the plurality of classifications.

11. The data classifier system according to claim 1, wherein the priority concerning a basic category X related to a plurality of classifications C in the classified hierarchy is calculated by

Priority (X:

C)=W1×

Independence (X;

C)+W2×

Specifics (X;

C) +W3×

Exhausitivity (X;

C)+W4×

Uniqueness (X;

C),

where W1, W2, W3and W4 denote weight coefficients to indexes which are determined in advance,

and wherein the Exhaustivity(X;

C) is given by

Exhaustivity (X;

C)=1/DataNum×

|U Data (ci)|,

where DataNum denotes an amount of data being classified;

Data(ci) denotes a set of data allocated to a classification ci selected from among the plurality of classifications; and

|U Data (ci)|denotes an amount of data allocated to the plurality of classifications.

12. The data classifier system according to claim 1, wherein the priority concerning a basic category X related to a plurality of classifications C in the classified hierarchy is calculated by

Priority (X:

C)=W1×

Independence (X;

C)+W2×

Specifics (X;

C) +W3×

Exhausitivity (X;

C)+W4×

Uniqueness (X;

C),

where W1, W2, W3, and W4 denote weight coefficients to indexes which are determined in advance,

and wherein the Uniqueness(X;

C) is given by

Uniqueness (X;

C)=1/(1/|U Data (ci) |×

CatNum(ci)),

where Data(ci) denotes a set of data allocated to a classification ci selected from among the plurality of classifications; and

| U Data (ci) | denotes an amount of data having no redundancy allocated to each of the plurality of classifications; and

CatNum(ci) denotes an amount of data allocated to the classification ci.

13. A data classifier method which refers to a classified hierarchy establishing a parent-child relationship between a plurality of classifications to produce a plurality of classification axes based on classifications and data groups input to the method, said data classifier method comprising:

a classification axis candidate reduction process which accumulates classifications serving as basic categories on which basis desired classifications in a database are selected, which selects a plurality of classifications from among classifications descendant from each basic category in consideration of semantic relevancy therebetween, to produce a plurality of classification axis candidates based on data quantity of classifications with reference to a common ancestor among classifications in the classified hierarchy; and

a priority calculation process which calculates priority in displaying the plurality of classification axis candidates with reference to hierarchical distances, each representing a length of a path reaching the common ancestor among classifications associated with classification axis candidates,

wherein one of;

(a) the priority concerning a basic category X related to a plurality of classifications C in the classified hierarchy is calculated by

Priority (X;

C)=1/Max(X)×

1/(2×

N)×

Σ

ComDist(ci,cj),

where Max(X) denotes a depth of a deepest classification among classifications descending from the basic category X;

N denotes a number of the plurality of classifications; and

ComDist(ci,cj) denotes a distance between classifications ci and cj selected from among the plurality of classifications; and

(b) the priority concerning a basic category X related to a plurality of classifications C in the classified hierarchy is calculated by

Priority (X;

C)=W1×

Independence (X;

C)+W2×

Specifics (X;

C) +W3×

Exhausitivity (X;

C)+W4×

Uniqueness (X;

C),

where W1, W2, W3 and W4denote weight coefficients to indexes which are determined in advance.

14. The data classifier method according to claim 13, wherein the classification axis candidate reduction process selects the plurality of classification axes with data quantity equal to or above a predetermined value or with highest and higher ranks of data quantities, wherein the plurality of classification axes is reduced within a specific range of lengths of paths reaching the common ancestor.

15. The data classifier method according to claim 13, wherein the priority calculation process calculates lengths of paths reaching the common ancestor in the classified hierarchy and determines priority for classification axis candidates based on the lengths of paths reaching the common ancestor, data quantity of classifications, or data redundancy of classifications.

16. A non-transitory computer-readable storage medium storing a data classifier program, executable by a computer including a processor and a storage unit, which refers to a classified hierarchy establishing a parent-child relationship between a plurality of classifications to produce a plurality of classification axes based on classifications and data groups input into the data classifier program, said data classifier program comprising:

a basic category accumulation process which accumulates classifications serving as basic categories on which basis desired classifications are selected;

a classification axis candidate reduction process which selects a plurality of classifications from among classifications descendant from each basic category in consideration of semantic relevancy therebetween, to produce a plurality of classification axis candidates based on data quantity of classifications with reference to a common ancestor among classifications in the classified hierarchy; and

a priority calculation process which calculates priority in displaying the plurality of classification axis candidates with reference to hierarchical distances, each representing a length of a path reaching the common ancestor among classifications associated with classification axis candidates,

wherein one of;

(a) the priority concerning a basic category X related to a plurality of classifications C in the classified hierarchy is calculated by

Priority (X;

C)=1/Max(X)×

1/(2×

N)×

Σ

ComDist(ci,cj),

where Max(X) denotes a depth of a deepest classification among classifications descending from the basic category X;

N denotes a number of the plurality of classifications; and

ComDist(ci,cj) denotes a distance between classifications ci and cj selected from among the plurality of classifications; and

(b) the priority concerning a basic category X related to a plurality of classifications C in the classified hierarchy is calculated by

Priority (X;

C)=W1×

Independence (X;

C)+W2×

Specifics (X;

C)+W3×

Exhausitivity (X;

C)+W4×

Uniqueness (X;

C),

where W1, W2, W3, and W4 denote weight coefficients to indexes which are determined in advance.

17. The non-transitory computer-readable storage medium according to claim 16, wherein the classification axis candidate reduction process produces the plurality of classification axis candidates with data quantity equal to or above a predetermined value or with highest and higher ranks of data quantities, wherein the plurality of classification axis candidates is reduced within a specific range of lengths of paths reaching the common ancestor.

18. The non-transitory computer-readable storage medium according to claim 16, wherein the data classifier program further comprises a display control process which performs display control based on the plurality of classification axis candidates output from the classification axis candidate reduction process and the priority e associated with data groups, wherein the display control process displays the plurality of classification axis candidates in an order of priority to allow a user to change classification axes or to select data groups for classification axes.

19. The non-transitory computer-readable storage medium according to claim 16, wherein the data classifier program further comprises a data retrieval process which retrieves data groups based on retrieval keywords to output retrieved data groups, wherein the classification axis candidate reduction process reduces the plurality of classification axis candidates based on the retrieved data groups, so that the priority calculation process calculates priority for classification axes correlated to the retrieved data groups.

20. The non-transitory computer-readable storage medium according to claim 16, wherein the data classifier program further comprises a data correlation process which establishes correlation between the classifications and the data groups.