The recommendation experiment we have conducted
in this paper demonstrates that it is effective to adjust
people’s mental health by providing recommendations.
Furthermore, our research also shows that recommender
system performs fairly well with high accuracy on mental
prediction.
Apparently, there exists vast space we can do to either
predict mental health status or help people alleviate mental
disorders. We are now exploring more efficient learning
algorithms to train better user model to predict mental
health. Meanwhile, we will try different methods to produce recommendations more effectively.