Development of a recommender system to choose a university degree program

• Main Author: Ler, Lian Ping
• Other Authors: Josephine Chong Leng Leng
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Computer and Information Science; Data mining
• Online Access: https://hdl.handle.net/10356/180880
• Institution: Nanyang Technological University

With the vast amount of information on the internet, users are increasingly struggling to find the information or items that cater to their interests or purpose at the moment. While information retrieval tools such as Google is available to narrow our search space, most results generated address the general population which might not be within our interests [1]. More specifically, when it comes to deciding a master’s degree, there are numerous factors to consider such as requirements, interests and prospects [2]. Hence, in this report, it aims to provide direction for people who are considering a master’s degree program, through the use of machine learning models and dashboard analytics. Hybrid switching model was proposed in this research incorporating K-Nearest Neighbours (KNN), Decision Tree (DCT) and Random Forest Classifier (RFC) as the foundation models. After the fine tuning of models, a desktop application was developed which provided the Graphical User Interface (GUI) for users to interact with the recommender system. The results of the model accuracy with other metrics were used to improve the features of the recommender system and to provide users with understanding on the outputs generated. Lastly, future works were proposed based on the limitations surfaced in this recommender system.