Predicting stress levels in humans accurately could offer numerous potential benefits in the clinical realm. However, conventional machine learning methodologies often fail to deliver high performance in this specific field. This shortfall is likely due to the unsuitability of a universal machine learning model in forecasting phenomena such as mood and stress, which display significant variability due to individual distinctions. Hence, in this repository, we leverage Multitask Learning (MTL) strategies to construct tailor-made machine learning models.
We facilitate this by categorizing individuals into clusters based on similar personality traits, then applying the multitask methodologies to these groups. We find the approach significantly effective.
Each section of this repository requires a seperate enviornment due to the use of a personalized versions of common libraries.
Mention the license under which the project is distributed.
Chen, M., Wang, Z., Zhao, Z., Zhang, W., Guo, X., Shen, J., ... & Ning, G. (2021, August). Task-wise split gradient boosting trees for multi-center diabetes prediction. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining (pp. 2663-2673).
Taylor, S., Jaques, N., Nosakhare, E., Sano, A., & Picard, R. (2017). Personalized multitask learning for predicting tomorrow's mood, stress, and health. IEEE Transactions on Affective Computing, 11(2), 200-213.