Detecting the ovarian cancer using big data analysis with effective model

Abstract

Now a day, a vast amount of medical data is available which can be efficiently utilized for diagnostic procedure by adopting data mining concepts. The main objective of this paper is to use effective data mining approach on huge amount of ovarian cancer dataset to identify the diseases in an efficient way. Thus, in this paper propose a novel approach for identifying ovarian cancer using combined Self-Organizing Maps Immune Clonal Selection (SOMICS) and Grammatical Evolution Neural Networks (GENN). SOMICS algorithm used for better feature selection which is used for extracting valuable, implicit, and interesting information from vast amount of medical data and GENN is used for classification process. The experimental results show the comparison of the proposed method and other classification methods using three various classifiers such as, Support Vector Machine (SVM), Multi-Layer Perceptron (MLP), Feed Forward Neural Network (FFNN). The combined SOMICS and GENN method yields promising results on classification and feature selection accuracy for ovarian cancer dataset with 98.23% classification accuracy, 0.0021% mean square error.