A Recommendation System for improving student life Based on Neural Network and Matrix Vectorization

Universities and schools have in recent years increased their efforts toward collecting and mining data related to students during their academic term. The amount of collected data is dedicated to studies for a better understanding of causalities, correlations, and effects of various inputs on student life and their academic results. In this work, a recommender system approach for students’ daily activities within their institution campuses is proposed. The aim is to recommend tasks to the student throughout the day in order to take some load off of the students when planning their daily activities. The recommender system is based on Artificial Neural Networks (ANNs) and Matrix factorization and is tested on the Dartmouth College StudentLife study dataset. The recommendation is validated using a portion of the dataset and results along with conclusions are drawn and presented.