Title: Enhancement of classification and prediction accuracy for breast cancer detection using fast convolution neural network with ensemble algorithm

Authors: Naga Deepti Ponnaganti; Raju Anitha

Addresses: Department of Computer Science and Engineering, KLEF, Vaddeswaram, India ' Department of Computer Science and Engineering, KLEF, Vaddeswaram, India

Abstract: Breast cancer is a leading cancer found mostly in women across the world and is more in number in the developing countries where they are not diagnosed in the early stages. The recent works compare machine learning algorithms using various techniques such as ensemble methods and classification. Hence, the requirement of time is to develop the technique which gives minimum error to increase accuracy. So, this paper proposes the neural network where classifying and predicting breast cancer is enhanced with maximum accuracy. Here, the novel technique of fast convolution neural network (FCNN) has been used for enhancing the classification and for improving the prediction accuracy ensemble algorithm of gradient boosting and adaptive boosting. By this proposed technique with ensemble algorithm, the huge data has been taken and predicted for detecting the cancer and this combined boosting algorithm will reduce the misclassification and will improve the binary classification. The training and testing of the dataset has been done with FCNN where the numerous datasets can be classified for earlier detection of cancer. The simulation result shows the improved accuracy, prediction class, precision and F-1 score.

Keywords: breast cancer; machine learning algorithms; classification; prediction accuracy; fast convolution neural network; FCNN; gradient boosting; adaptive boosting.

DOI: 10.1504/IJCSE.2023.129735

International Journal of Computational Science and Engineering, 2023 Vol.26 No.2, pp.171 - 181

Received: 01 Feb 2021
Accepted: 01 Nov 2021

Published online: 22 Mar 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article