Abstract
With the emphasis on revitalization of small and medium-sized enterprises, the importance of supporting systems for consultation of small and medium-sized enterprises is increasing, which supports small and medium-sized enterprises to solve problems that are difficult to deal with by themselves. Consulting companies can respond to a wide range of management consultations but service proposal and problem detection depend on the experience and intuition of the consultant since the content of consultation is wide and highly specialized, so that it is difficult to provide a stable service. A support system for providing stable services independent of the expertise of consulting company is desired. In this research, as first step in constructing a support system, by analyzing text data describing the content of consultation with client companies, a discriminant formula to predict future problems is created. In this paper, termination problem is targeted and comparative experiments is conducted to verify the effectiveness of the proposed method.
