Indian Journal of Science and Technology, Vol 9(11), DOI: 10.17485/ijst/2016/v9i11/67151, March 2016 ISSN (Print) : 0974-6846 ISSN (Online) : 0974-5645 * Author for correspondence Abstract The objective of this paper is to analyze and identify the best classification solution for clinical decision making. Several classification algorithms Like Discriminant Analysis (LDA), Support Vector Machine (SVM), Artificial Neural Network (ANN), Naive Bayes (NB), and Decision Trees are compared to find the optimum diagnostic accuracy. The performance of classification algorithms are compared using benchmark dataset, breast cancer. The effects of normalization using z-score and min-max approaches are also investigated. The results are compared based on different performance parameters like accuracy, sensitivity, specificity and root node error value. Accuracy has been improved for all classifications methods after normalizing the data set. Z-score normalization performs better for all the measures when compared to min-max normalization. The proposed approach shows higher accuracy rate for Naive Bayes algorithm when compared with the other algorithms. Keywords: Accuracy, Classification, Min-Max Normalization, Sensitivity, Specificity, Z-Score Normalization Analytical Study of Selected Classification Algorithms for Clinical Dataset V. R. Suma, Shwetha Renjith, Sreeja Ashok * and M. V. Judy Department of Computer Science and IT, Amrita School of Arts and Sciences, Amrita Vishwa Vidya peetham, Kochi - 682024, Kerala, India; sumavrillam92@gmail.com, shweshweranjith@gmail.com, sreeja.ashok@gmail.com, judy.nair@gmail.com 1. Introduction Data mining is an analysis step in KDD process to investigate pre-existing huge database in order to create new information that is understandable for future use. DM has applied in many different areas, i.e., Financial Data Analysis, Retail Industry, Telecommunication Industry, Biological Data analysis, Scientific Applications, Medical Domain, Intrusion Detection etc. Data mining is used in medical field because it is used todeterminefresh trends, significant patterns from data. Classification, Clustering, Association etc., are the functionalities of data mining.Classificationisa supervised learning process that classifies dataand analyses and extracts representation for data classes. ese models are called classifiers. ese classifiers predicts the truthful class label. Classification can be explained as a two-stepprocedure 1 : Step 1: Learning Step-Predict a classification model. Training data set is analyzed by classification algorithms and the classifiers are derived in the form of classification rules. Step 2: Classification Step-Model is used to predict class labels for new dataset. e accuracy of the classifiers is identified with the percentage of valid data set that is rightly classified by the classifier. For that we use a Confusion matrix (Contingency Table). It is a square matrix where each feature represents instance of a predicted class and each line represents instance of authentic class. It visually represents the performance of an algorithm 1 . In this paper five efficient classification techniques such as Linear Discriminant Analysis (LDA), Support Vector Machine (SVM), Naive Bayes (NB), Artificial Neural Network (ANN), and Decision Tree (DT) are consider for comparing and evaluating the performance. Section 2 deals with explanation of different classification algorithms. Section 3 deals with advantages and disadvantages of classification methods. Section 4 explains the results and discussionandSectionconcludes with final remarks.