Machine Learning Algorithms for the Prediction of Prostate Cancer

Abstract

Machine learning (ML) algorithms employ a wide variety of methods such as statistical, probabilistic and optimization that allow machines to learn and predict based on historical data. For these capabilities, ML is substantially used in different cancer prediction. The exact prediction is essential for the planning of cancer treatment. This work aims at exploring the idea of using ML algorithms to predict prostate cancer. To increase the probability of survival of prostate cancer patients, it is important to establish suitable prediction models. Therefore, in this study, we applied several ML techniques including support vector machine, k-nearest neighbors, Naive Bayes, random forest and logistic regression algorithms to predict prostate cancer. Among all the five ML techniques, the logistic regression provided better prediction result with 86.21% accuracy. Hence, our achieved results indicate that the logistic regression technique could be utilized for prostate cancer prediction.