Method of analyzing conversational transcripts

1. A computer-implemented method for analyzing transcripts of conversation between at least two users, the method comprising:

receiving, by a computer, input information from a first user;

creating, by said computer, a conversational transcript from said input information received from said first user;

selecting, by said computer, a defined situation from a list of defined situations to filter said conversational transcript;

comparing, by said computer, said defined situation with knowledge derived from historical conversational transcripts to identify a set of procedural sequences, said comparing comprising:

clustering, by said computer, textual segments in said historical conversational transcripts into groups of related textual segments;

representing, by said computer, each of said groups of related textual segments by a representative syntax;

forming, by said computer, a sequence of said representative syntax for each conversational transcript;

finding, by said computer, frequently occurring subsequences from said sequence of representative syntax; and

finding, by said computer, distinct and long, frequently occurring subsequences from the frequently occurring subsequences that are identified with said defined situation;

identifying, by said computer, said defined situation in said conversational transcript;

identifying, by said computer, a set of procedural sequences by comparing said defined situation in said conversational transcript with knowledge derived from said historical conversational transcripts; and

providing, by said computer, said set of procedural sequences to said first user.

2. The method of claim 1 , said knowledge derived from said historical conversational transcripts comprising:

clustering textual segments in said historical conversational transcripts into groups of related textual segments;

representing each of said groups of related textual segments by a representative syntax;

forming a sequence of said representative syntax for each historical conversational transcript;

finding frequently occurring subsequences from said sequence of representative syntax; and

finding distinct and long frequently occurring subsequences from said frequently occurring subsequences.

3. The method of claim 2 , said clustering textual segments comprising:

organizing said historical conversational transcripts into groups of conversational transcripts according to a defined criteria; and

clustering said groups of conversational transcripts to obtain said groups of related textual segments.

4. The method of claim 3 , said clustering said groups of conversational transcripts further comprising:

generating at least two sets of documents from said groups of conversational transcripts.

5. The method of claim 2 , each said sequence of representative syntax comprises a segment of conversational transcripts belonging to each of said groups of related textual segments.

6. The method of claim 2 , said sequence of representative syntax comprising a set of frequent and distinct words in each of related textual segments.

7. The method of claim 2 , said finding distinct and long frequently occurring sub-sequences being performed by a clustering algorithm belonging to the class of leader clustering algorithm.

8. The method of claim 2 , said clustering textual segments in said historical conversational transcripts into groups of related textual segments further comprising utilizing a set of configurable clustering parameters to cluster said related textual segments.

9. The method of claim 1 , further comprising:

prompting a second user on appropriate procedural sequences based on said defined situation, said second user providing said appropriate procedural sequence to said first user.