Abstract: D-matrix is a systematic diagnostic model which is used to catch the fault system data and its causal relationship at the hierarchical system-level. Developing a D-matrix from first standards and updating it using the domain information is a labor intensive and time consuming task. Further, in-time augmentation of D-matrix through the discovery of new symptoms and failure modes observed for the first time is a challenging task. Is describes construction and updation of D-Matrix by automatically mining the unstructured repair verbatim ( written in unstructured text) data collected during fault diagnosis using document annotation, term extraction and phrase merging. The system construct the fault diagnosis ontology consisting of concepts and relationships commonly observed in the fault diagnosis domain. Next, employ the text mining algorithms make use of ontology concept to identify the necessary artifacts, such as parts, symptoms, failure modes, and their dependencies from the unstructured repair verbatim text. The method is implemented as a prototype tool and validated by using real-life data collected from the automobile domain.

Keywords: Data Mining, fault analysis, fault diagnosis, information retrieval, text processing.