Diagnosis of medical dataset using fuzzy-rough ordered weighted average classification

Now-a-days more deadly diseases are affecting human life and the major concern is prediction of such diseases. The Proposed methods is to classify the disease from the available dataset with four Phases : Data collection and preprocessing, Fuzzy Rough Feature Selection, Ordered weighted average classification, Performance analysis. In preprocessing, Kernel filters converts the given set of prediction variables into a kernel matrix. The attribute values remains unchanged, as long as the preprocessing kernel filters doesn't change it. Feature Selection is a process which attempts to select features which are more informative. It is done using Fuzzy rough set which utilizes a fuzzy rough dependency measure to eliminate redundancy features in a backward elimination may be a search strategy. Fuzzy rough ordered weighed average classification is used to predict group membership for data instances. The goal of classification is to accurately predict the target class for each case in the data. Finally the predicted model is measured using different performance analysis like accuracy, sensitivity, specificity and kappa statistics. The resulting classification accuracy provides a better accuracy than the existing methods with minimum features found using fuzzy rough set.