Classification of Factors Affecting Renal Failure by Machine Learning Methods

44900

Machine learning methods are widely used for data analysis in health research. The aim of this study is to classify the factors that affect renal failure by using various machine learning methods such as Artificial Neural Networks (Multilayer Perceptron), Support Vector Machines, Naive Bayes, Decision Trees, Random Forests, K-Nearest Neighborhood algorithms. In this study, 237 patients who have been in emergency unit in Hospital of Numune in Ankara and were older than 18 years and have upper gastrointestinal bleeding symptoms have been selected. Here, 34 variables such as age, gender, blood values, other diseases etc. which affect renal failure have been used to make classification with machine learning methods. When machine learning methods are compared according to the accuracy rates, F-measure, sensivity, specifity and Kappa values, it has been found that decision trees algorithm performs well.

88

Machine Learning , Classification , Decision Trees , Renal Failure

Erciyes University Journal Of The Institute Of Science an‎d Technology

101

Erciyes University Journal Of The Institute Of Science an‎d Technology