The Implementation of Neural Network Algorithm to Predict The Eligibility of Prospective Assistants

E-learning has been able to store learning data accumulated from the beginning of students entering college in the initial semester to the final semester. The problems occurring are that each semester requires an assistant for each course with practical activities, and there are difficulties in determining for prospective assistants who have excellent abilities. Almost every course in the Department of Informatics Engineering courses has a teaching assistant. In each semester, the courses will always open vacancies for the registration of teaching assistants. The selection should refer to the data from the learning outcomes to measure the feasibility of prospective applicants. To obtain information, data processing was required, frequently referred to as data mining. The stages of this study began with searching for literature studies, selecting data mining methods following the desired results, collecting data, processing data with rapidminer devices, applying and testing models with neural network algorithms, discovering findings, and drawing conclusions. The data used in this study were e-learning quiz data, data availability, and students' final grades. The attributes employed were the average score of the quiz, attendance, final grade, and description. Attribute information was utilized as a label. In this study, the authors obtained an accuracy of 98.18%. In conclusion, the neural network algorithm was excellent in predicting the prospective assistant lecturers.