Recommender System with Machine Learning and Artificial Intelligence

7Machine Learning-Based Recommender System for Breast Cancer Prognosis

G. Kanimozhi, P. Shanmugavadivu* and M. Mary Shanthi Rani

Department of Computer Science and Applications, The Gandhigram Rural Institute (Deemed to be University), Gandhigram, Tamil Nadu, India

Abstract

The prognosis of the onset of cancer plays an inevitable role in saving the lives of the victims. The proposed “Machine Learning based Recommender System for Breast Cancer Prediction (MLRS-BC)” aims to provide an accurate recommendation for breast cancer prognosis through four distinct phases, namely: Data collection; Preprocessing; Training, Testing, Validation; and Prediction/Recommender. It is designed to predict the effect of risk factors associated with routine blood analysis in the Breast Cancer Coimbra Dataset (BCCD). The attributes of BCCD are age, body mass index, glucose, and insulin level in the blood, Homa, Leptin, Adiponectin, Resistin, and Monocyte Chemoattractant Protein-1 (MCP-1). The Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) is used to evaluate the accuracy of the predictions. The MLRS-BC computes the error values for each attribute of BCCD. It recommends the best attribute having the least error rate as the pre-dominant attributes for breast cancer prognosis. It gains importance in automated breast cancer detection or classification, with a single optimal attribute, instead of engaging all the nine attributes of the dataset. MLRS-BC also recommends the best prediction algorithm ...