Predicting Mental Health Status in the Context of Web Browsing

Currently, people around the world are suffering from mental disorders. Given the wide-spread use of the Internet, we propose to predict users´ mental health status based on browsing behavior, and further recommend suggestions for adjustment. To identify mental health status, we extract the user´s web browsing behavior, and train a Support Vector Machine (SVM) model for prediction. Based on the predicted status, our recommender system generates suggestions for adjusting mental disorders. We have implemented a system named Web Mind as the experimental platform integrated with the predicting model and recommendation engine. We have conducted user study to test the effectiveness of the predicting model, and the result demonstrates that the recommender system performs fairly well.