Development of an machine learning-based system for determining the vocational future of students who will transfer to higher education

The aim of this study is to develop a machine learning-based system that can predict the vocational future of students who will transfer to higher education by determining the relationship between students' interests, abilities, personality traits, and departments. The study was carried out with 410 students studying at the Faculty of Education of a university in the Eastern Anatolia Region in the spring semester of the 2020-2021 academic year. The research was based on the survey model. The researcher used, value questionnaires, interest, talent, personality inventory, and personal information form developed by the researcher were used as data collection tools. The data was collected through Google Form and analyzed through both statistical and machine learning methods. As a result of the findings obtained through statistical methods, it has been determined that there is a relationship between the departments of the students participating in the study and the types of values, interests, and talents and that their personality traits don't differ according to the faculty departments. When the performances of the models trained through artificial learning methods were examined, it was determined that the decision tree model had an accuracy of 77%, linear regression of 48%, and logistic regression of 40%. The R-value for the model trained in artificial neural networks was determined as 0.83. To determine the student's vocational future a web-based system has been developed through machine learning methods. Due to the presence of categorical data in the data set, the model was trained through logistic regression and a system was created that can predict the occupational field according to the demographic and psychological characteristics of the students through the Python programming language and the PyCharm program.