Summarization Techniques in Association Rule Data Mining For Risk Assessment of Diabetes Mellitus
Abstract— At Early exposure of patients with dignified risk of developing diabetes mellitus is so hyper critical to the bettered prevention and global clinical management of these patients. In an existing system, apriori algorithm is used to find the itemsets for association rules but it is not efficient in finding itemsets and it uses only four association rules for finding the risk of diabetes mellitus so it have low precision. In this paper we are focusing to implement association rule mining to electronic medical records to detect set of danger factors and their equivalent or identical subpopulations that indicates patients at especially steep risk of progressing diabetes. Association rule mining accomplishes a very bulky set of rules for summarizing the EMR with huge dimensionability. We proposed a system in enlargement to combine risk of diabetes for the purpose of finding an suitable summary for this we use ten association rule and using the reorder algorithm for finding the itemsets and rules. For identifying the risk we considered four association rule set summarization techniques and organised a related calculation to support counselling with respect to their applicability merits and demerits and provide solutions to reduce the risk of diabetes. The above four methods having its fair strength but the bus algorithm developed the best acceptable summary.
Index Terms— Data Mining; Fuzzy Clustering means Algorithm; Association rule mining; Association rule summarization techniques.
Click Here
International Journal for Trends in Technology & Engineering © 2015 IJTET JOURNAL