Question and answer data editing device, question and answer data editing method and question answer data editing program

• US 8,983,962 B2
• Filed: 02/08/2006
• Issued: 03/17/2015
• Est. Priority Date: 02/08/2005
• Status: Active Grant

• Current Assignee: NEC Corporation
• Original Assignee: NEC Corporation
• Inventors: Nakazawa, Satoshi, Ikeda, Takahiro, Satoh, Kenji
• Primary Examiner(s): SOMERS, MARC S
• Application Number:
  US11/815,621
• Publication Number:
  US 20080195378A1
• Time in Patent Office: 3,324 Days
• Field of Search: 707/705, 707/706, 707/722, 707/723, 707/725, 707/726, 707/728, 707/731, 707/736, 707/737, 707/738, 707/740, 707/741, 707/758, 707/769, 707/771, 707/776, 707/780, 707/917, 707/944, 705/304
• US Class Current: 707/738
• CPC Class Codes: G06F 16/3329 Natural language query formulation or dialogue systems
• CPC Class Codes: Y10S 707/917 Text
• CPC Class Codes: Y10S 707/944 Business related

Abstract

The question and answer data editing device for editing dialog content to generate question and answer data, includes a detecting unit that detects a part of the dialog content similar to existing question and answer data stored, and a extracting unit that extracts a context in which the dialog content is made from dialog content in the proximity of the similar part detected and registers the context extracted as new question and answer data or as index information of the question and answer data.

Claims

1. A question and answer data editing device for editing the content of a dialogue to generate question and answer data, comprising:

a processor configured to function as at least one of a matching unit and a question and answer variation extraction unit,

wherein the matching unit detects a first question part or a first answer part from a history data of the content of the dialogue as being similar to a first question and answer data in existing question and answer data; and

the question and answer variation extraction unit extracts;

a) a second answer part from the history data, when a second question part in the history data is similar to the first question and answer data and the second answer part in the history data is not similar to the first question and answer data, and;

b) a third question part from the history data, when a third answer part in the history data is similar to the first question and answer data and the third question part is not similar to the first question and answer data; and

registers said third question part or said second answer part from said content of the dialogue as a variation of said first question and answer data in the existing question and answer data.

2. The question and answer data editing device according to claim 1, further comprising a question and answer data editing unit that presents said detected first question part or said first answer part in said content of the dialogue or a portion of said first question part and said first answer part so as to be edited in correlation with said existing question and answer data.

6. The question and answer data editing device according to any one of claims 1 to 5 further comprising:

a dialogue history data base for recording the history data of said dialogue content,

a question and answer data base for recording existing question and answer data, and

a matching condition input unit that receives input of conditions of said dialogue content and first question and answer data to be processed, wherein

said dialogue content and said question and answer data coincident with the input conditions to be processed are read from said dialogue history data base and said question and answer data base.

3. A question and answer data editing device for editing the content of a dialogue to generate question and answer data, comprising:

a processor configured to function as at least one of a matching unit and an associated question and answer extraction unit, wherein the matching unit detects a first question part or a first answer part from said dialogue content similar to a first question and answer data from existing question and answer data; and

the associated question and answer extraction unit extracts;

a) a second answer part from a history data, when a second question part in the history data is similar to the first question and answer data and the second answer part in the history data is not similar to the first question and answer data, and;

b) a third question part from the history data, when a third answer part in the history data is similar to the first question and answer data and the third question part is not similar to the first question and answer data; and

registers the third question part or the second answer part as associated question and answer data of said first question and answer data,

wherein the third question part or second answer part adjacent to the first question part or said first answer part in the content of the dialogue.

4. The question and answer data editing device according to claim 3, further comprising a question and answer data editing unit that presents said extracted associated question and answer data so as to be edited in correlation with said existing question and answer data.

5. A question and answer data editing device for editing the content of a dialogue to generate question and answer data, comprising:

a processor configured to function as at least one of a matching unit, an expression pattern extraction unit, a question and answer variation extraction unit and an associated question and answer variation extraction unit,

wherein the matching unit detects a first question part or a first answer part from a history data of said dialogue content that is similar to a first question and answer data included in existing question and answer data;

the expression pattern extraction unit extracts an expression pattern including a context or a condition in which said dialogue was made from the proximity of said first question part or said first answer part, and registers said extracted expression pattern as index information of said first question and answer data;

the question and answer variation extraction unit extracts a second question part or a second answer part not similar to said first question and answer data from the proximity of said first question part or said first answer part and registers said extracted second question part or second answer part as a variation of said first question and answer data;

the associated question and answer extraction unit extracts;

a) a third answer part from the history data, when a third question part in the history data is similar to the first question and answer data and the third answer part in the history data is not similar to the first question and answer data, and;

b) the third question part from the history data, when the third answer part in the history data is similar to the first question and answer data and the third question part is not similar to the first question and answer data, and;

registers said third question part or said third answer part as associated question and answer data of said first question and answer data.

18. A dialogue supporting device for supporting a dialogue by using said generated question and answer data or said index information which is generated by the question and answer data editing device according to claim 5.

7. A question and answer data editing method of editing the content of a dialogue to generate question and answer data, comprising:

detecting a first question part or a first answer part from a history data of the content of the dialogue as being similar to a first question and answer data in existing question and answer data;

extracting a) a second answer part from the history data, when a second question part in the history data is similar to the first question and answer data and the second answer part in the history data is not similar to the first question and answer data, and b) a third question part from the history data, when a third answer part in the history data is similar to the first question and answer data and the third question part is not similar to the first question and answer data; and

registering said third question part or said second answer part from said content of the dialogue as a variation of said first question and answer data.

8. The question and answer data editing method according to claim 7, further comprising a step of presenting said extracted question and answer data in said content of the dialogue or a part of said extracted question and answer data so as to be edited in correlation with said existing question and answer data extracted.

9. A question and answer data editing method of editing the content of a dialogue to generate question and answer data, comprising:

detecting a first question part or a first answer part from history of said dialogue content similar to a first question and answer data from existing question and answer data;

extracting a second question and answer data from the existing question and answer data; and

registering the first question and answer data and the second question and answer data as associated question and answer data,

wherein the registering occurs in response to a detection of a second question part or second answer part similar to said second question and answer data, the second question part or second answer part adjacent to the first question part or said first answer part in the content of the dialogue.

10. The question and answer data editing method according to claim 9, further comprising a step of presenting said associated question and answer data extracted so as to be edited in correlation with said existing question and answer data.

12. The question and answer data editing method according to any one of claims 7 to 11 further comprising:

receiving input of conditions of said dialogue content and first question and answer data to be read from a dialogue history data base for recording history data of said dialogue content and a question and answer data base for recording existing question and answer data and to be processed, and

reading said dialogue content and said question and answer data coincident with the input conditions from said dialogue history data base and said question and answer data base.

11. A question and answer data editing method of editing the content of a dialogue to generate question and answer data, comprising:

detecting a first question part or a first answer part from a history data of said dialogue content that is similar to a first question and answer data included in existing question and answer data;

extracting an expression pattern including a context or a condition in which said dialogue was made from the proximity of said first question part or said first answer part;

registering said extracted expression pattern as index information of said first question and answer data;

extracting a) a second answer part from the history data, when the second question part in the history data is similar to the first question and answer data and the second answer part in the history data is not similar to the first question and answer data, and b) a third question part from the history data, when a third answer part in the history data is similar to the first question and answer data and the second question part is not similar to the first question and answer data;

registering said extracted third question part or second answer part as a variation of said first question and answer data; and

extracting a second question and answer data from the history data of said dialogue content as associated question and answer data to said first question and answer data in response to detecting that a third question part or a third answer part similar to the extracted second question and answer data is present in the vicinity of the first question part or the first answer part.

19. A dialogue supporting method of supporting a dialogue by using said generated question and answer data or said index information which is generated by the question and answer data editing method according to claim 11.

13. A non-transitory computer readable medium storing a question and answer data editing program for causing a computer to execute a method of editing content of a dialogue to generate question and answer data, the method comprising:

detecting a first question part or a first answer part from a history data of the content of the dialogue as being similar to a first question and answer data in existing question and answer data;

extracting a) a second answer part from the history data, when a second question part in the history data is similar to the first question and answer data and a second answer part in the history data is not similar to the first question and answer data, and b) a third question part from the history data, when a third answer part in the history data is similar to the first question and answer data and the third question part is not similar to the first question and answer data; and

registering said third question part or said second answer part from said content of the dialogue as a variation of said first question and answer data.

14. The non-transitory computer readable medium according to claim 13, wherein

said question and answer data editing program further comprising a function of presenting said extracted question and answer data in said content of the dialogue or a part of said extracted question and answer data so as to be edited in correlation with said existing question and answer data extracted.

17. The non-transitory computer readable medium according to any one of claims 13 to 16, wherein first question and answer data editing program further comprising the functions of:

receiving input of conditions of said dialogue content and said question and answer data to be read from a dialogue history data base for recording history data of said dialogue content and a question and answer data base for recording existing question and answer data and to be processed, and

reading said dialogue content and said question and answer data coincident with the input conditions from said dialogue history data base and said question and answer data base.

15. A non-transitory computer readable medium storing a question and answer data editing program for causing a computer to execute a method of editing content of a dialogue to generate question and answer data, said method comprising:

detecting a first question part or a first answer part from history of said dialogue content similar to a first question and answer data from existing question and answer data;

extracting a second question and answer data from the existing question and answer data; and

registering the first question and answer data and the second question and answer data as associated question and answer data,

wherein the registering occurs in response to a detection of a second question part or second answer part similar to said second question and answer data, the second question part or second answer part adjacent to the first question part or said first answer part in the content of the dialogue.

16. A non-transitory computer readable medium storing a question and answer data editing program for causing a computer to execute a method of editing the content of a dialogue to generate question and answer data, said method comprising:

detecting a first question part or a first answer part from a history data of said dialogue content that is similar to a first question and answer data included in existing question and answer data;

extracting an expression pattern including a context or a condition in which said dialogue was made from the proximity of said first question part or said first answer part;

registering said extracted expression pattern as index information of said first question and answer data;

extracting a) a second answer part from the history data, when a second question part in the history data is similar to the first question and answer data and a second answer part in the history data is not similar to the first question and answer data, and b) a third question part from the history data, when a third answer part in the history data is similar to the first question and answer data and the third question part is not similar to the first question and answer data;

registering said extracted third question part or second answer part as a variation of said first question and answer data; and

extracting a second question and answer data from the history data of said dialogue content as associated question and answer data to said first question and answer data in response to detecting that a third question part or a third answer part similar to the extracted second question and answer data is present in the vicinity of the first question part or the first answer part.

20. A dialogue supporting non-transitory computer readable medium for causing said computer to execute a function of supporting a dialogue by using said generated question and answer data or said index information which is generated by the question and answer data editing program according to claim 16.