Abstract:

In this study, it is aimed to examine the theses written in the field of machine learning and decision trees with the bibliometric analysis method by evaluating them within the scope of various parameters. For this purpose, 368 theses were reached as a result of the search performed in May 2020 by using the keywords "machine learning", "decision trees" in the database of the Council of Higher Education's National Thesis Center. Theses reached; It was examined in detail in terms of bibliography taking into account the year, thesis type, language, university, institute, department, the number of pages, the software used, the methods used, the keywords. In view of the research questions defined for the theses, the data were obtained and transferred to the Excel program. In this software, all calculations, graphs and tables were made. As a result of the examinations, important findings were obtained regarding the theses written in the fields of machine learning and decision tree based on the supervised learning strategy. These include important findings such as the distribution of theses by years, the university, institute, department that contributed the most to the field, the most commonly used machine learning method, the feature selection methods used, the parameter optimization methods used, the hybrid approaches developed, the ensemble learning methods developed, the most popular programming language / software, etc. Key Words: Bibliometric Analysis, Thesis, Machine Learning, Supervised Learning, Decision Trees JEL Classification: C38

Keywords: