Investigation of machine learning classification algorithms

In this study, it is aimed to present in a clearer and clear with the application of Decision Tree, Naive Bayes, Random Forest and K-nearest neighboring classification algorithms which are seen to have a complex theory within the literature, to different data types with the help of codes written in R program. In addition, a performance comparison of the above-mentioned four machine learning classification algorithms has been made with the help of the KNIME program, which has recently come to the forefront with end-user-friendly innovations for machine learning methods over a real data set. In the first section of the study, an introduction was made and in the second part, basic concepts about big data, machine learning, performance evaluation methods, classification and classification success measures were mentioned. In the following chapters, mathematical concepts for Decision Tree, Naive Bayes, Random Forest and K-nearest neighboring classification algorithms are given and the importance of using these methods is emphasized. In the fourth part of the study, using different data types, R applications of classification algorithms discussed in previous chapters and a performance comparison in KNIME is made over a real dataset. In the last section, the results of the study are discussed. Key Words: Classification Algorithms, Machine Learning, Big Data.