Fibonacci polynomial based multilayer perceptron neural network for classification of medical data

Now-a-days, neural networks have appeared as prominent field for medical data classification. Where, Chebyshev Multilayer perceptron (CMLP) neural network is commonly used for the classification. Meanwhile, this algorithm generates the enhanced inputs with large values using Chebyshev polynomial, which increase the computational complexity of the network. To overcome this problem, Fibonacci polynomials based MLP is proposed to generate the enhanced inputs with the smaller value to reduce the computational complexity with more accuracy. The results of the proposed algorithm are compared with some statistical and machine learning algorithms like Narve Bayes, Bayes net, Ada Boost, SVM (support vector machine) and MLP (multilayer perceptron). Six medical datasets are collected from the UCI and KEEL data repositories for the classification task. The comparative analysis of proposed and comparison algorithms show that the proposed algorithm performs well as compared to the others in terms of Accuracy Sensitivity, specificity, precision, f-measure, and Area under the curve.