PERFORMANCE EVALUATION OF ALGORITHMS FOR BREAST CANCER PREDICTION

  • Khushbu Verma, M. P. Thapliyal
Keywords: Breast cancer; Machine Learning; Prediction Techniques; Deep learning.

Abstract

Breast cancer is a cause of a large number of deaths of women around the world. The prediction of medical disease can be efficiently done by machine learning techniques, for early prediction of breast cancer, this paper presents machine learning algorithms such as Logistic Regression, Random Forest, Sequential Minimal Optimization, and proposes a Deep learning algorithm. In this paper, we have proposed a deep learning algorithm that obtained optimal results using the hyper-parameter techniques. The dataset applied on experimental tool revealed that our proposed deep neural network (DNN) algorithm is 98% accurate for the prediction of breast cancer dataset, whereas the existing algorithm shows the accuracy up to 96%, 97% and 96% with respect to Logistic Regression (LR), Random Forest (RF) and Sequential Minimal Optimization (SMO), respectively. Therefore, we conclude that the proposed deep neural network (DNN) algorithm leads to the maximum accuracy up to 98% by using the hyper-parameter techniques.

Published
2021-09-17
How to Cite
M. P. Thapliyal, K. V. (2021). PERFORMANCE EVALUATION OF ALGORITHMS FOR BREAST CANCER PREDICTION. Design Engineering, 8184-8193. Retrieved from http://www.thedesignengineering.com/index.php/DE/article/view/4445
Section
Articles