Deep learning for NTUQA forum question classification

• Main Author: Fung, Joseph King Yiu
• Other Authors: Hui Siu Cheung
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2021
• Subjects: Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence
• Online Access: https://hdl.handle.net/10356/153054
• Institution: Nanyang Technological University

NTU Question and Answer (NTUQA) forum is a web-based application that allows users to post Past Year Paper (PYP) questions, share answers and upvote or downvote answers. NTUQA was first developed by Project Officer Nguyen Quang Sang using the Django framework based Askbot open-source software. This project aims to introduce a tag recommendation feature and offer better visualization of all the questions in the database separated into different clusters to find inconspicuous concepts between similar questions. These two enhancements are essentially multi-label text classification and clustering tasks in Natural Language Processing (NLP). This report describes all the steps of the research beginning from the related research to the implementation and then to evaluation.