International Journal of Electrical and Computer Engineering (IJECE) Vol. 8, No. 6, December 2018, pp. 4505~4518 ISSN: 2088-8708, DOI: 10.11591/ijece.v8i6.pp4505-4518  4505 Journal homepage: http://iaescore.com/journals/index.php/IJECE SCDT: FC-NNC-structured Complex Decision Technique for Gene Analysis Using Fuzzy Cluster based Nearest Neighbor Classifier Sudha V. 1 , Girijamma H. A. 2 1 Department of IS&E, RNS Institute of Technology, India 2 Department of Computer Science & Engineering, RNS Institute of Technology, India Article Info ABSTRACT Article history: Received Feb 12, 2018 Revised Jun 17, 2018 Accepted Jun 20, 2018 In many diseases classification an accurate gene analysis is needed, for which selection of most informative genes is very important and it require a technique of decision in complex context of ambiguity. The traditional methods include for selecting most significant gene includes some of the statistical analysis namely 2-Sample-T-test (2STT), Entropy, Signal to Noise Ratio (SNR). This paper evaluates gene selection and classification on the basis of accurate gene selection using structured complex decision technique (SCDT) and classifies it using fuzzy cluster based nearest neighborclassifier (FC-NNC). The effectiveness of the proposed SCDT and FC-NNC is evaluated for leave one out cross validation metric(LOOCV) along with sensitivity, specificity, precision and F1-score with four different classifiers namely 1) Radial Basis Function (RBF), 2) Multi-layer perception(MLP), 3) Feed Forward(FF) and 4) Support vector machine(SVM) for three different datasets of DLBCL, Leukemia and Prostate tumor. The proposed SCDT &FC-NNC exhibits superior result for being considered more accurate decision mechanism. Keyword: Fuzzy classification Gene analysis Gene selection Machine learning Micro array data Copyright © 2018 Institute of Advanced Engineering and Science. All rights reserved. Corresponding Author: Sudha V., Department of Information Science & Engineering, RNS Institute of Technology, Bengaluru, India. Email: sudhavinayakam@gmail.com 1. INTRODUCTION The accuracy of diagnosis is the basis for the perfect treatment process to be adopted especially in the case of fatal disease like cancers, leukemia and prostrate tumor etc. Along with the histopathology, medical radiology and imaging techniques, the micro-array data analysis could be proven quite helpful as well as rightful if efficient techniques of analysis are evolved [1]. The accuracy of disease classification or early diagnosis depends upon, how accurately the gene of significance is selected. The DNA-microarray data analysis is challenging in both aspects of statistically and computationally as it possesses non-linear noises along with high dimensionality of low sample data [2]. Many efforts towards disease diagnosis particularly cancer, tumor etc, classification have been seen in literature [3]-[10]. The section 2 describes the insights of related work. Various machine learning approaches are used for the classification which includes radial basis function (RBF), artificial neural network (ANN), support vector machine (SVM) etc. by forming the problem as binary classification. The problem of dimension reduction for searching most significant gene is being formulated as many problem spaces which includes 1) Mixed integer programming (MIP), 2) Bio-inspired optimization (BIO), 3) Mining association rules (MAR), and last but not the least 4) Ensemble technique (ET) [8]. The clinically comprehensive method requires handling high dimensional data with veracity and noises to handle ambiguity during the right gene candidate selection. This paper proposes a mechanism of