Method of analyzing conversational transcripts

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.