AI for social good

• Main Author: Lim, Yuh Horng
• Other Authors: Erik Cambria
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Engineering::Computer science and engineering
• Online Access: https://hdl.handle.net/10356/166073
• Institution: Nanyang Technological University

Suicide is a prevalent global issue that affects people of all ages and regions, with 703,000 people taking their own life every year. Suicide is a tragic event that not only impacts individuals but also has lasting effects on their loved ones. Globally, suicide is the fourth leading cause of death among 15–29-year-olds. With the recent development in machine learning and prevalence of social media, it warrants a further exploration in this problem. This project explores the use of machine learning to aid in suicide prevention efforts by analyzing data from popular social media platforms, Twitter and Reddit. Various techniques such as CNN, LSTM, GRU, and Transfer Learning were implemented to identify text with suicidal ideation. Additionally, a Twitter bot was developed to detect the presence of suicidal ideation and provide support to those in need. Limitations of the project, such as the difficulty of identifying humor and sarcasm in text, are discussed, and potential future directions, including multi-task learning and chatbot-based psychological assessments, are proposed. Ultimately, this project highlights the importance of early intervention in preventing suicide and demonstrates the potential of machine learning in suicide prevention efforts.