Application for Selection of Student Final Project Supervisors Based on the Selected Category and Expertise of Lecturers Using the Naive Bayes Classifier Method

Abstract

At the end of the task the supervisor has an important role for the success achieved and graduation of students. For this reason, ideal supervisors are needed for students. As discussed in the STMIK Hang Tuah Pekanbaru in the process of submitting the title of this thesis, so are some of the problems that arise, namely regarding the matter of coaching because the process is still using conventional methods that is based on personal knowledge of the Head of Study Program, the difficulty of the development process of submitting the Student's final position for difficulties check the final supervisor's assignment. The application of selecting the final project supervisor for students is the solution of the debate. The supervisor lecturer recommendation system that can utilize the naïve bayes classifier algorithm as a determinant of the probability of the lecturer results students can choose. Naive Bayes is a prediction technique based on simple probabilities based on the application of the Bayes theorem (Bayes rule) with a strong assumption of independence. The selection is based on the final criteria for the assignment and expertise of the lecturer. From the application of this recommendation is obtained from the recommendations of supervisors in accordance with the concept of the student's final project. With reference data, training and Bayes rules obtained sufficient results to satisfy students in getting a supervisor who is in accordance with the topic of the student's final project.