Research On Life Style Disease Knowledge Acquisition Based On Rough Set

The intelligent detection of life style disease is concerned by many countries around the world, but because of complexity of the symptoms of life style disease, it is difficult to get a precise accurate mathematical model and limit its research. In recent years, people have paid more and more attention to the study on life style disease, and established the reasoning system based on expert experience, and the experience of experts has some subjectivity and information is often incomplete and inaccurate. As a new method, rough set has been well applied in practice. In this paper, a general method for knowledge acquisition and reasoning based on rough set is proposed. The method of knowledge acquisition and life style disease detection are systematically and deeply studied:The construction of knowledge base includes: data acquisition, data preprocessing, attribute reduction. In this paper, attribute reduction was mainly studied. Rough set reduction algorithm based on genetic algorithm is used to reduce the redundant attributes of the original data, reduce the size of the rule base, using the reduction algorithm to initialize the population, the initial population corresponding position gene is limited, so that it has some decision table reduction results attribute characteristics, avoid the blindness of the population, reduce the search space, and accuracy of the results is greatly improved. According to the support degree calculated by the reduction results, the certain medical knowledge base and the certain medical knowledge base was established to improve the effectiveness of the construction of knowledge base.On the basis of summarizing the research of others, an improved reasoning method based on the importance of attribute was proposed and was applied in reasoning detection. The method has no need of any prior experience, the obtaining of the importance of attribute has no need of prior experience, and it is completely obtained from the sample space of the domain, which avoids the subjectivity, and the certain medical knowledge base and the certain medical knowledge base was introduced, through the parameter K weight distribution, considering the importance of each rule, adjust the output to get effective reasoning of disease probability.At last, the process of building the knowledge base was described by heart disease data and verified the proposed reasoning method. The results show that the application of modeling method based on rough set theory has adaptability to the detection of heart disease, which can meet the needs of practical application. According to the results of this study, this method has a good sharing, and can be applied to other complexity of the symptoms of life style disease like diabetes, high blood pressure and so on.