The earlier work on the
.The earlier work on the competitor mining utilized the text data to collect comparative evidences between two items. But, the comparative evidences are based on the assumptions, which may not always exist. Competitor identification is referred to as a classification process through which competitors of a focal firm are identified based on “relevant similarities”. Authors in developed an automatic system that discovers competing companies from public information sources. In this system data is crawled from text and it uses transformation oriented learning to obtain appropriate data normalization, modeling to represent models of linked data, and Bayesian network for competitor identification technique is used.