Predicting Recurrent Chat Contact In A Psychological Intervention For The Youth Using Natural Language Processing

This article discusses the use of Natural Language Processing (NLP) in predicting recurrent contact for chat-based counseling hotlines for youth mental health. The study utilized a large text corpus and an XGBoost Classifier to achieve an AUROC score of 0.68 in predicting whether chatters would contact the service again. The results showed that words indicating younger age or female gender and terms related to self-harm and suicidal thoughts were associated with a higher chance of recontacting. This approach could lead to personalized care for young people in need of mental health support.