ABSTRACT

Detection of knowledge patterns in clinicial data through data mining. Data mining algorithms can be trained from past examples in clinical data and model the frequent times non-linear relationships between the independent and dependent variables. The consequential model represents formal knowledge, which can often make available a good analytic judgment. Classification is the generally used technique in medical data mining. This paper presents results comparison of ten supervised data mining algorithms using five performance criteria. We evaluate the performance for C4.5, SVM, K-NN, PNN, BLR, MLR, CRT, CS-CRT, PLS-DA and PLS-LDA then Comparison a performance of data mining algorithms based on computing time, precision value , the data evaluated using 10 fold Cross Validation error rate, error rate focuses True Positive, True Negative, False Positive and False Negative, bootstrap validation and accuracy. A typical confusion matrix is furthermore displayed for quick check. The study describes algorithmic discussion of the dataset for the disease acquired from UCI, on line repository of large datasets. The Best results are achieved by using Tanagra tool. Tanagra is data mining matching set. The accuracy is calculate based on addition of true positive and true negative followed by the division of all possibilities.