Sentic computing for HIV prevention and care

• Main Author: Low, Valerian Qin Ling
• Other Authors: Erik Cambria
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2022
• Subjects: Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence
• Online Access: https://hdl.handle.net/10356/156348
• Institution: Nanyang Technological University

The Action for AIDS (AFA) has established a community roadmap towards eradicating AIDS and HIV in Singapore by 2030. However, the results in Singapore now, despite being almost achieving the “90-90-90” goals set by the UNAIDS, still fall short of achieving the target that “90% of people living with HIV (PLHIV) will be aware of their HIV status”. In this paper, we will be performing sentiment analysis on interviews with PLHIV, which allows us to extract the sentiment polarity and emotions, thus being able to better understand the current HIV situation in Singapore. We have developed a complete sentiment analysis system that automatically processes the data and stores the results. We used the APIs available from SenticNet and attempted to improve the accuracy using Multi-Task Learning.