Development of Knowledge Base in Expert System using Dempster's Rule of Combination¹

Expert system has been used in many fields such as medicine, agriculture, forestry, public health, economics etc. Expert system represent the expert to provide advice to the user so that no longer need to meet directly with an expert. Knowledge base is the most important and decisive part in expert system. Errors in the knowledge base would result in an incorrect conclusion. However, due to time and budget constraints, an expert will be difficult to gain knowledge from a variety of sources including research then the information derived from the public can be an important source of knowledge for expert system. In this paper we purpose a model for development of a knowledge base in an expert system by combining expert knowledge and information or evidence from the public (tacit knowledge) that expected an expert system that is always factual. The combination of evidence using Dempster's rule of combination.