Data mining project

Task Yeast rpart Dataset: • Description: [login to view URL] • Data: [login to view URL]: Objective: To predict the localization site of protein – to find the best classifier according to the precision for CYT class and overall accuracy measures. The original classes should be replaced by 4 classes {CYT, NUC, (ME1, ME2, ME3 ), rest). A script should include: • Data analysis. • A description of an idea for looking for the best classifier. • Information about carried out experiments (parameters of a method, a confusion matrix, a value of the accuracy measure for test data and the precision measure for the CYT class). • Conclusions referring to obtained results. Only one split into test and training data should be used. Classifiers should be built by using rpart function.