A Deep Neural Network in a Web-based Career Track Recommender System for Lower Secondary Education

Abstract

In this paper, web-based career track recommender system was used to guide guidance counselor in assisting their students in choosing an appropriate career track. Many junior high school students struggled with track uncertainty and were perplexed when it came to deciding whether senior high school career track was appropriate and suitable for them. Increased in drop-out rate is also a bigger concern in the country, and students switching to another program can be a waste of government funds intended for free tuition at state universities. Given the current state of K-12 evaluation, adequate counseling of guidance counselor in the selection of relevant career tracks should be undertaken. This study included 1500 students from the first to third grades of the K-12 curriculum, and their grades and socio-demographic profiles were used as factors in determining their academic strand in Senior High School with the utilization of Deep Neural Network. The study's findings suggest that the DNN algorithm predicts the academic strand of students quite well with a prediction accuracy of 83.11%. Using the devised approach, guidance counselors' work became more efficient in dealing with student concerns. With the use of the DNN technique, it is concluded that the recommender system acts as a decision tool for counselors in advising their students to select which Senior High School track is appropriate for them. The web-based career track recommender system has effectively integrated the DNN predictive model.