期刊名称：International Journal of Innovative Research in Science, Engineering and Technology
印刷版ISSN：2347-6710
电子版ISSN：2319-8753
出版年度：2013
卷号：2
期号：6
页码：2157
出版社：S&S Publications
摘要：Association rules are an energetic investigating area. Association rules characterize a promisingmethod to search syndrome differentiation on modern India. Solitary of the most accepted approach to do datamining is determining association rules. The association innovation is an imperative research field in datamining. The mining association rule frequently has been adopts numerous models: support, confidence,interestingness. But this model can’t accurate measure the correlative degree between the precursor and theconsequential of the rule by allocation. So we proposed a new mining model of association rules: support,coincidence, interestingness and investigate the significance of fluke by instance. We use this model in the dataabout coronary heart disease and obtained a lot of meaningful rules. Proposed a new model of supportcoincidence-interestingness base on the traditional model of support-confidence interestingness. Our proposemodel can quantitatively evaluate the correlation of rules and reduce many rules that have low support or haveno correlation or have negative correlation. In our work we will conduct experiments on large real time topredict the diseases like Medication in Coronary Heart Disease and compare the performance of our algorithmwith other related algorithms. Our propose model based on CMAR (Classification based on MultipleAssociation Rules) SVM, fuzzy discriminant Analysis.
关键词：Coronary Heart Disease; SVM; fuzzy Discriminant fuzzy diminishing support
