Design of Ensemble Classifier Model Based on MLP Neural Network For Breast Cancer Diagnosis

Nowadays, breast cancer is one of the leading causes of death women in the worldwide. Substantial support for breast cancer awareness and research funding has helped created advances in the diagnosis and treatment of breast cancer. Data mining techniques have a growing reputation in the medical field because of high diagnostic capability and useful classification and they can help breast cancer diagnosis. In this paper, a Multi-Layer Perceptron Neural Network (MLP-NN) based on Evolutionary Algorithms (EA) is used to automatically classify breast cancer. Here, EA is used to tune MLP parameters such as optimal features, hidden layers, hidden nodes and weights. Ensemble models is a machine learning approach to combine multiple other single models in the prediction process. To improve the performance of the classification model, we use an Intelligent Ensemble Classification method based on MLP, named IEC-MLP. The proposed method was evaluated for the samples of the Wisconsin Breast Cancer Dataset (WBCD) by stacked generalization technique. The proposed method was evaluated for the samples of the Wisconsin database by stacked generalization technique. Experimental results show the advanced performance of the IEC-MLP with ensemble classifiers compared to other algorithms. Accordingly, IEC-MLP was better than GAANN and CAFS algorithms with classification accuracy of 98.74%.

Automated summary

This paper utilized a Multi-Layer Perceptron Neural Network (MLP-NN) based on Evolutionary Algorithms (EA) to automatically classify breast cancer, and evaluated its performance using stacked generalization technique. Experimental results demonstrated the superior performance of IEC-MLP with ensemble classifiers compared to other algorithms.