A machine learning algorithm for expert system based on subjective Bayesian method

Based on subjective Bayesian method, knowledge representation model and inference method were analyzed, and the formula for certainty in reasoning process was given, and what lead to the inaccurate sufficient factor and necessary factor were discussed, and a correction algorithm to adjust the sufficient factor and the necessary factor to the true one based on case studies was given in this paper. Experiments proved that even the sufficient factor and necessary factor were inaccurate at the beginning, by using the correction algorithm, the sufficient factor and necessary factor could converge at actual values after adequate number of case studies.