Comparative Analysis of Data Reduction Model for Diabetes

Mining of the Data now a day plays a major role and concern in the present world in the industry and also in the research areas. -Data mining is the process of extracting hidden information from a large set of database and it can help researchers gain both novel and deep insights of unprecedented understanding of large biomedical datasets. Data mining can uncover new biomedical and healthcare knowledge for clinical decision making. Many tools and application software’s are evolved to mine the data and to present the best data visualization. One of the Data Mining tool available is WEKA, in which contain many machine level and other types of algorithms. The WEKA software was employed as mining tool for diagnosing diabetes. The Pima Indian diabetes database (PIDD) was obtained from UCI server and used for analysis. The dataset consist a total instance of 768 and 9 attributes (8 for input and 1 for output) which will be used to test and justify the differences between different filtering, attribute selection and classification methods. On the basis of relative absolute error, Stratified Remove Folds, Resample performed equally and these two outperformed Randomize, Remove Folds and Sparse to Non-Sparse. And on the basis of mean absolute error, Randomize and Sparse to Non-Sparse outperformed other three, and they both perform equally well. KeywordsData Mining, KDD, healthcare sector, Classification, diabetes.