Constructing Depression Prediction Model using ChatGPT and Machine Learning Algorithms

Abstract

Depression is currently one of the main causes of self-harm in human society. However, the insufficient diagnosis of depression has been a long-standing problem. Traditional diagnostic methods heavily rely on the patient's current emotions, and patients often hesitate to seek help, which frequently delays intervention. With the deep expansion of social media, individuals with suicidal thoughts often express their perspectives and thoughts on these platforms. Relevant studies have also found that people with depression are more likely to disclose their condition. Therefore, there is a potential to enhance the identification of users at risk of suicide by analyzing social media posts. Although machine learning has been successfully applied to build depression prediction models, the brevity of social media posts often affects the classification performance of these models. Hence, this study aims to introduce ChatGPT to automatically generate experimental corpus using information from short comments. Natural language processing (NLP), Support Vector Machines (SVM), and Naive Bayes (NB) classifiers are then employed to enhance the classification accuracy of the depression prediction model. The experimental results validate the effectiveness of the proposed method.