Method and system for providing electronic discovery on computer databases and archives using statement analysis to detect false statements and recover relevant data

• US 7,366,714 B2
• Filed: 12/06/2004
• Issued: 04/29/2008
• Est. Priority Date: 03/23/2000
• Status: Expired due to Term

• Current Assignee: Albert B. Krachman
• Original Assignee: Albert B. Krachman
• Inventors: Krachman, Albert
• Primary Examiner(s): Al-Hashemi; Sana
• Assistant Examiner(s): Jami; Hares
• Application Number:
  US11/005,213
• Publication Number:
  US 20050086226A1
• Time in Patent Office: 1,240 Days
• Field of Search: 707/3, 707/1
• US Class Current: 1/1
• CPC Class Codes: G06Q 50/16 Real estate
• CPC Class Codes: G06Q 50/26 Government or public services business processes related to the transportation industry G06Q50/40

Abstract

Electronic discovery on computer systems and archives is provided by using statement analysis of to identify false statements so as to retrieve relevant impeachment data and other evidence related to recorded statements. The recorded statement is analyzed for possible false statements and corresponding transcript data of the false statement, any precipitating question, and any follow-up questions are selected for use in the query. The selected text can be further processed before being used in the query. The query can be used in a search engine or used to train smart search agents. The search engine query or smart search agents are released onto target computer systems and/or archives to search for responsive data and documents. Notification, reports, and indexing of responsive data and documents can be provided to produce relevant results. The analysis of the recorded statement is preferably done by voice stress analysis.

Claims

1. A method for providing electronic discovery on computer databases and archives using artificial intelligence to produce smart search agents to locate relevant impeachment data and data related to a potentially false statement comprising:

inputting information relevant to desired data related to an issue into a neural network;

training the neural network to produce a search algorithm in the form of a smart search agent, wherein said training comprises a query produced by analyzing recorded statements for potential false statements comprising analyzing an audio portion of the recorded statements using a voice stress algorithm and selecting transcript data corresponding to said potentially false statements as input for the query;

selecting a target computer database and archives to search for responsive data and documents;

searching the target computer databases and archives using said smart search agent; and

outputting results responsive to the searching.

2. The method for providing electronic discovery on computer databases and archives of claim 1, wherein said voice stress analysis is performed by a software module on a computer.

3. The method for providing electronic discovery on computer databases and archives of claim 1, wherein said issue is a legal issue and said relevant data are legally relevant documents.

4. The method for providing electronic discovery on computer databases and archives of claim 3, further comprising indexing the said results for review of possible confidential and/or privileged material.

5. The method for providing electronic discovery on computer databases and archives of claim 4, further comprising locking the results when a user does no have appropriate access privileges and unlocking the results when a user has appropriate access privileges.

6. The method for providing electronic discovery on computer databases and archives of claim 1, further comprising scanning paper documents into the computer database.

7. The method for providing electronic discovery on computer databases and archives of claim 1, further comprising inputting the target database as information relevant to desired data for training the neural network.

8. The method for providing electronic discovery on computer databases and archives of claim 1, wherein the neural network is a legal concept neural network.

9. The method for providing electronic discovery on computer databases and archives of claim 1, wherein the discovery is for a purpose selected from the group consisting of litigation, alternative dispute resolution, security clearance, regulatory compliance, legal risk management, and criminal law enforcement.

10. The method for providing electronic discovery on computer databases and archives of claim 1, wherein the analyzing recorded statements is selected from the group consisting of voice stress analysis, polygraph analysis, and human-detected lies.

11. The method for providing electronic discovery on computer databases and archives of claim 1, wherein the selecting transcript data corresponding to said potentially false statements comprises selecting at least a portion of transcript text of the false statement, any preceding question, and any follow-up question.

12. The method for providing electronic discovery on computer databases and archives of claim 11, further comprising processing the selected transcript text by a process selected from the group consisting of word deletion, word weighting, synonym addition, and antonym addition.

13. A system for providing electronic discovery on computer databases and archives using artificial intelligence to produce smart search agents, comprising:

means for inputting information relevant to desired data related to an issue into a neural network;

means for training the neural network to produce a search algorithm in the form of a smart search agent, wherein said means for training further comprises means to input a query produced by a means for analyzing an audio portion of recorded statements for potential false statements with a voice stress analyzer and selecting transcript data corresponding to said potentially false statements;

means for searching a target computer database and archive using the smart search agent; and

means for outputting results responsive to said means for searching.

14. The system for providing electronic discovery on computer databases and archives of claim 13, wherein said voice stress analyzer is a computer-based software module.

15. The system for providing electronic discovery on computer databases and archives of claim 13, wherein said issue is a legal issue.

16. The system for providing electronic discovery on computer databases and archives of claim 15, wherein said means for outputting results comprises means for indexing the results in a form capable of review for possible confidential and/or privileged material.

17. The system for providing electronic discovery on computer databases and archives of claim 16, wherein said means for outputting results comprises means to lock a file against users not having appropriate privileges and to unlock file a file for users having appropriate privileges.

18. The system for providing electronic discovery on computer databases and archives of claim 13, further comprising means for manually selecting any part of the transcript and using the selected part as a query for databases.

19. The system for providing electronic discovery on computer databases and archives of claim 13, wherein said means for selecting transcript data corresponding to the potentially false statements comprises means for selecting at least a portion of transcript text comprising the false statement, any preceding question, and any follow-up question.

20. The system for providing electronic discovery on computer databases and archives of claim 19, further comprising means for processing the selected transcript text selected from the group consisting of means for word deletion, means for word weighting, means for synonym addition, and means for antonym addition.

21. A process of using statement analysis and artificial intelligence to find impeachment data related to a recorded statement, comprising:

inputting appropriate data related to said recorded statement into an artificial intelligence system;

developing a search algorithm using the artificial intelligence system;

converting the search algorithm into a smart search agent, wherein the smart search agent comprises a query produced by analyzing an audio portion of the recorded statement with a voice stress analyzer for potential false statements, and selecting transcript data corresponding to said potentially false statements as input for said query; and

deploying said smart search agent on relevant computer systems to identify electronic material relevant to impeach the recorded statement.

22. The process of using statement analysis and artificial intelligence to find impeachment data related to a recorded statement of claim 21, further comprising indexing identified material for review of possible confidential and/or privileged material.

23. The process of using statement analysis and artificial intelligence to find impeachment data related to a recorded statement of claim 21, wherein said recorded statement is selected from the group consisting of depositions, court testimony, legal hearing testimony, regulatory hearing testimony, legislative processes, interviews and data produced in discovery.

24. The process of using statement analysis and artificial intelligence to find impeachment data related to a recorded statement of claim 21, further comprising locking the identified material when a user does no have appropriate access privileges and unlocking the identified material when a user has appropriate access privileges.

25. The process of using statement analysis and artificial intelligence to find impeachment data related to a recorded statement of claim 21, further comprising selecting any part of a transcript and using the selected part as a query for databases.

26. The process of using statement analysis and artificial intelligence to find impeachment data related to a recorded statement of claim 21, wherein said selecting transcript data corresponding to said potentially false statements comprises selecting at least a portion of transcript text of the false statement, any preceding question, and any follow-up question.

27. The process of using statement analysis and artificial intelligence to find impeachment data related to a recorded statement of claim 26, further comprising processing the selected transcript text by a process selected from the group consisting of word deletion, word weighting, synonym addition, and antonym addition.

28. A system for providing electronic discovery on computer databases and archives using a search engine, comprising:

means to input a query to a search engine, wherein said search query is produced by a means for analyzing an audio portion of recorded statements with a voice stress analyzer for potential false statements and selecting transcript data corresponding to said potentially false statements;

means for searching a selected target computer database and archive using the search engine; and

means for outputting results responsive to the means for searching.

29. The system for providing electronic discovery on computer databases and archives of claim 28, wherein the voice stress analyzer is a computer-based software module.

30. The system for providing electronic discovery on computer databases and archives of claim 28, wherein said search engine comprises means for training a neural network.

31. The system for providing electronic discovery on computer databases and archives of claim 28, wherein the means for outputting results comprises means for indexing the results in a form capable of review for possible confidential and/or privileged material.

32. The method for providing electronic discovery on computer databases and archives of claim 31, wherein the means for indexing comprises means for locking the results when a user does not have appropriate access privileges and unlocking the results when a user has appropriate access privileges.

33. The system for providing electronic discovery on computer databases and archives of claim 28, further comprising means for selecting any part of the transcript and using the selected part as a query for databases.

34. The system for providing electronic discovery on computer databases and archives of claim 28, wherein said means for selecting transcript data corresponding to said potentially false statements comprises means for selecting at least a portion of transcript text of the false statement, any preceding question, and any follow-up question.

35. The system for providing electronic discovery on computer databases and archives of claim 34, further comprising means for processing the selected transcript text selected from the group consisting of means for word deletion, means for word weighting, means for synonym addition, and means for antonym addition.

36. A process of using statement analysis and a search engine to find impeachment data related to an audio portion of a recorded statement, comprising:

inputting an appropriate query related to said audio portion of said recorded statement into a search engine, wherein said query is produced by analyzing the audio portion of said recorded statement by voice stress analysis for potential false statements and selecting transcript data corresponding to said potentially false statements as input for said query; and

deploying said search engine on relevant computer systems to identify electronic material relevant to impeach said recorded statement.

37. The process of using statement analysis and a search engine to find impeachment data related to a recorded statement of claim 36, further comprising indexing identified material for review of possible confidential and/or privileged material.

38. The process of using statement analysis and a search engine to find impeachment data related to a recorded statement of claim 36, wherein the recorded statement is selected from the group consisting of depositions, court testimony, legal hearing testimony, regulatory hearing testimony, legislative processes, interviews and data produced in discovery.

39. The process of using statement analysis and a search engine to find impeachment data related to a recorded statement of claim 37, further comprising locking the identified material when a user does not have appropriate access privileges and unlocking the identified material when a user has appropriate access privleges.

40. The process of using statement analysis and a search engine to find impeachment data related to a recorded statement of claim 36, further comprising selecting a part of the transcript and using the selected part as a query for databases.

41. The process of using statement analysis and a search engine to find impeachment data related to a recorded statement of claim 36, wherein the search engine comprises artificial intelligence and the process comprises training of a neural network.

42. The process of using statement analysis and a search engine to find impeachment data related to a recorded statement of claim 36, wherein said selecting transcript data corresponding to said potentially false statements comprises selecting at least a portion of transcript text of the false statement, any preceding question, and any follow-up question.

43. The process of using statement analysis and a search engine to find impeachment data related to a recorded statement of claim 42, further comprising processing the selected transcript text by a process selected from the group consisting of word deletion, word weighting, synonym addition, and antonym addition.