A Survey of Machine Learning Techniques for Cancer Disease Prediction and Diagnosis Kumar M Kiran1,*, Udayan J Divya2 1School of Computer Science and Engineering, VIT, Vellore, India 2School of Information Technology and Engineering, VIT, Vellore, India *Corresponding author: M. Kiran Kumar, E-mail: kiran9sai@gmail.com
Online published on 6 April, 2019. Abstract Identification of patterns plays a vital role in Disease Diagnosis for detecting the diseases accurately. Machine learning is a subfield of artificial intelligence (AI). This ML techniques are mostly interesting as it is part of a suggesting personalized, predictive medicine to the diseases. Cancer is the one of the second leading cause of death worldwide, in 2018 it is accountable for an estimated 9.7 million deaths. The most common cancers are breast, oral, skin, colon and lung. Cancer death will be reduced if cases are identified and treated early. In connection with this review a broad survey is conducted for various machine learning algorithms used for prediction and prognosis of different cancer diseases. A number of methods are reviewed, towards various types of cancer diseases, a heavy reliance on “historical” technologies used in machine learning methods. At the end the benefits & limitations & challenges are identified which helps the researches to develop novel methodologies in ML to improve the performance in disease prediction and diagnosis. Top Keywords Machine learning, Medical diagnosis, Classification algorithms, Decision trees, KNN, ANN, SVM. Top |