Application Of Machine Learning Algorithm In Medical Data Analysis

In recent years, with the rapid development of information technology, computer technology and the Internet, various social fields accumulate a large scale of data. Then the traditional statistical analysis methods has exposed its limitation. How to excavate useful information in such large-scale data has become very significant problems of all works of life.Machine learning is one of the main methods to solve the problems of data mining. Machine learning is a process of self-improvement by the system itself,so that a computer program can automatically improve performance with accumulated experience. Although these algorithms are not enough to make the machine think like humans, but they have made a great breakthrough in many areas. Especially in the medical data, the machine learning algorithms have showed great superiority.This paper mainly introduces four kinds of machine learning algorithms, namely k nearest neighbor, decision tree, support vector machine and random forest, and apply the four algorithms in chronic kidney disease data to establish respective statistical classification models by optimizing the parameters, and then, we compare the mistake sentencing rate of these four algorithms on this medical data by cross validation, and found the mistake sentencing rate of these four kinds of machine learning algorithms is very low. At the same time, the mistake sentencing rate of the random forest is the lowest, it is only 0.0025, so in the last. We chose random forest model to analyze and forecast the data of chronic kidney disease(CKD).