Methodology to establish term co-relationship using sentence boundary detection

A method for using sentence boundaries to establish term co-relationships for fault model development, said method comprising:

• providing a document corpus containing service technician text verbatim records to be analyzed;
• analyzing the document corpus to identify sentence boundaries and produce a list of sentences;
• extracting terms and establishing term correlations from the list of sentences, where the terms include parts, symptoms, failure modes, and service actions;
• and
• validating the term correlations to produce a validated term correlation database.

A system for establishing term co-relationships using sentence boundaries, said system comprising:

• a rule base containing rules for identifying sentence-ending periods;
• a sentence boundary detection module for applying the rules in the rule base to text records contained in a document corpus, and producing a list of sentences;
• an extraction and correlation module for extracting terms from the list of sentences and establishing correlations between the terms based on sentence boundaries;
• and
• a validation module for validating the correlations between the terms based on frequency of occurrence.