Method and apparatus for capturing and presenting digital data in a synthetic interview

• US 5,870,755 A
• Filed: 02/26/1997
• Issued: 02/09/1999
• Est. Priority Date: 02/26/1997
• Status: Expired due to Term

• Current Assignee: Carnegie Mellon University
• Original Assignee: Carnegie Mellon University
• Inventors: Stevens, Scott M., Christel, Michael G.
• Primary Examiner(s): Downs, Robert W.
• Application Number:
  US08/806,926
• Time in Patent Office: 713 Days
• Field of Search: 707/104, 706/59, 706/45
• US Class Current: 1/1
• CPC Class Codes: G06F 16/7844 using original textual content or text extracted from visual content or transcript of audio data
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing
• CPC Class Codes: Y10S 707/99948 Application of database or data structure, e.g. distributed, multimedia, or image

Abstract

A method of creating a database for facilitating a synthetic interview begins with the step of generating questions. The questions and responses to the questions are recorded. A transcript of the recorded questions and responses is created. An association between each question and the response thereto is maintained. Each of the responses and each of the questions is expanded with semantic information. Inverted indices are created for the semantic expansions of the responses, the questions, and the transcripts of the responses and questions to enable retrieval of the recorded responses in a seamless manner. A method is also disclosed for creating a database for creating a synthetic interview from existing material.

Claims

1. A computer implemented method of processing recorded data for facilitating a synthetic interview therefrom, comprising the steps of:

creating a transcript of the recorded data;

identifying statements in the recorded data;

associating a question with each of said statements;

expanding each of said statements and questions with semantic information; and

creating inverted indices of said questions, said statements, said semantic expansions of said statements, and said semantic expansions of said questions to enable retrieval of the statements.

2. A method of creating a database for facilitating a synthetic interview, comprising the steps of:

generating questions;

recording said questions and the responses to said questions;

creating a transcript of the recorded information;

maintaining an association between each question and said response thereto;

expanding said responses and said questions with semantic information;

creating inverted indices of said questions, said responses, said semantic expansions of said questions, and said semantic expansions of said responses; and

storing said inverted indices and said recorded information in a database.

3. The method of claim 2 additionally comprising the step of recording generic pool statements for addition to the database.

4. A method of retrieving answers from a database in response to a question, said database containing indexed questions, answers responsive to said questions, and semantic expansions of said questions and answers, comprising the steps of:

receiving a query;

searching the indexed questions for matches and associated weights;

searching the indexed answers for matches and associated weights;

searching the indexed semantic expansion of the questions for matches and associated weights;

searching the indexed semantic expansion of the answers for matches and associated weights;

identifying the best match from the question search, answer search, and semantic expansion searches; and

displaying the answer associated with the best match.

5. The method of claim 4 additionally comprising the step of preprocessing the received query to extract stop words.

6. The method of claim 4 additionally comprising the step of determining if predetermined criteria are satisfied.

7. The method of claim 6 wherein said database includes pool statements, and wherein said method additionally comprises the step of displaying a pool statement if the predetermined criteria are not satisfied.

8. The method of claim 6 additionally comprising the steps of tracking received queries and best matches therefor, and updating the criteria based on the tracked information.

9. A computer readable medium encoded with a computer program, said computer program for processing recorded data for facilitating a synthetic interview therefrom, said computer program comprising a series of steps for:

creating a transcript of the recorded data;

identifying statements in the recorded data;

associating a question with each of said statements;

expanding each of said statements and questions with semantic information; and

creating inverted indices of said questions, said statements, said semantic expansions of said statements, and said semantic expansions of said questions to enable retrieval of the statements.

10. A computer readable medium encoded with a computer program, said computer program for creating a database for facilitating a synthetic interview, said computer program comprising a series of steps for:

creating a transcript of the recorded information;

maintaining an association between each question and the response thereto;

expanding said responses and said questions with semantic information;

creating inverted indices of said questions, said responses, said semantic expansions of said questions, and said semantic expansions of said responses; and

storing said inverted indices and said recorded information in a database.

11. A computer readable medium encoded with a computer program, said computer program for retrieving answers from a database in response to a question, said database containing indexed questions, answers responsive to said questions, and semantic expansions of said questions and answers, said computer program comprising a series of steps for:

receiving a query;

searching the indexed questions for matches and associated weights;

searching the indexed answers for matches and associated weights;

searching the indexed semantic expansion of the questions for matches and associated weights;

searching the indexed semantic expansion of the answers for matches and associated weights;

identifying the best match from the question search, answer search, and semantic expansion searches; and

displaying the answer associated with the best match.

12. The media of claim 11 wherein said computer program additionally comprises a step for preprocessing the received query to extract stop words.

13. The media of claim 11 wherein said computer program additionally comprises a step for determining if predetermined criteria are met.

14. The media of claim 13 wherein said data base includes pool statements, and wherein said computer program additionally comprises a step for displaying a pool statement if the predetermined criteria are not satisfied.

15. The media of claim 13 wherein said computer program additionally comprises steps for tracking received queries and best matches therefor, and updating the criteria based on the tracked information.

16. An apparatus for retrieving answers from a database in response to a question, said database containing indexed questions, answers responsive to said questions, and semantic expansions of said questions and answers, said apparatus comprising:

means for receiving a query;

means, responsive to said means for receiving, for searching the indexed questions, the indexed answers, the indexed semantic expansion of the questions, and the indexed semantic expansion of the answers for matches and associated weights;

means, responsive to said means for searching, for selecting the best answer to the question; and

means for displaying the best answer.

17. The apparatus of claim 16 additionally comprising means for preprocessing the received query to extract stop words.

18. The apparatus of claim 16 additionally comprising means for determining if predetermined criteria are satisfied.

19. The apparatus of claim 18 wherein said database includes pool statements, said apparatus additionally comprising means for displaying one of said pool statements if the predetermined criteria are not satisfied.

20. The apparatus of claim 18 additionally comprising means for tracking queries and best matches therefor, and means for updating the criteria based on the tracked information.