A Comparison of Machine Learning Techniques for Survival Prediction in Breast Cancer Gene Expression Data
8 Pages Posted: 27 Feb 2018
Date Written: November 15, 2017
The myriad developments Microarrays and the ability to perform massively parallel Gene Expression analysis of human tumours were received with great excitement by the scientific community. Machine learning algorithms are utilized in computer programs that try to predict in the (behaviour, cancer type and what an image is of, stock market fluctuations). It is based on what situations caused what results in the past. The eventual goal of machine learning in breast cancer diagnosis is given the gene expression levels or other data from a cancer patient, can accurately predict the level and severity of cancer that they have and aiding the doctors in treating them. This paper is a comparison of several different machine learning algorithms, relating their performance on diagnosing breast cancer from gene expression level data. The classifying breast cancer dataset, responsible a cancer has a mutation in the breast cancer gene expression dataset. To compares different machine learning algorithms: Decision Tree, K-nearest-neighbor,Naive Bayes Initially, I guessed that the Decision Tree would perform best simply because it is a very widespread machine learning algorithm.
Keywords: Machine Learning, Decision Tree (DT), Support Vector Machine (SVM), Naive Bayes, Microarray data, Breast cancer Gene Expression.
Suggested Citation: Suggested Citation