TY - GEN
T1 - Improving students' daily life stress forecasting using lstm neural networks
AU - Umematsu, Terumi
AU - Sano, Akane
AU - Taylor, Sara
AU - Picard, Rosalind W.
N1 - Funding Information:
This work is supported by National Institute of Health (R01GM105018), the MIT Media Lab Consortium, NEC Corporation and Samsung Electronics.
Publisher Copyright:
© 2019 IEEE.
PY - 2019/5
Y1 - 2019/5
N2 - Accurately forecasting stress may enable people to make behavioral changes that could improve their future health. For example, accurate stress forecasting might inspire people to make changes to their schedule to get more sleep or exercise, in order to reduce excessive stress tomorrow night. In this paper, we examine how accurately the previous N-days of multi-modal data can forecast tomorrow evening's high/low binary stress levels using long short-Term memory neural network models (LSTM), logistic regression (LR), and support vector machines (SVM). Using a total of 2,276 days, with 1,231 overlapping 8-day sequences of data from 142 participants (including physiological signals, mobile phone usage, location, and behavioral surveys), we find the LSTM significantly outperforms LR and SVM with the best results reaching 83.6% using 7 days of prior data. Using time-series models improves the forecasting of stress even when considering only subsets of the multi-modal data set, e.g., using only physiology data. In particular, the LSTM model reaches 81.4% accuracy using only objective and passive data, i.e., not including subjective reports from a daily survey.
AB - Accurately forecasting stress may enable people to make behavioral changes that could improve their future health. For example, accurate stress forecasting might inspire people to make changes to their schedule to get more sleep or exercise, in order to reduce excessive stress tomorrow night. In this paper, we examine how accurately the previous N-days of multi-modal data can forecast tomorrow evening's high/low binary stress levels using long short-Term memory neural network models (LSTM), logistic regression (LR), and support vector machines (SVM). Using a total of 2,276 days, with 1,231 overlapping 8-day sequences of data from 142 participants (including physiological signals, mobile phone usage, location, and behavioral surveys), we find the LSTM significantly outperforms LR and SVM with the best results reaching 83.6% using 7 days of prior data. Using time-series models improves the forecasting of stress even when considering only subsets of the multi-modal data set, e.g., using only physiology data. In particular, the LSTM model reaches 81.4% accuracy using only objective and passive data, i.e., not including subjective reports from a daily survey.
KW - Forecasting
KW - LSTM
KW - Objective
KW - Stress
KW - Wearable
UR - http://www.scopus.com/inward/record.url?scp=85073035877&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85073035877&partnerID=8YFLogxK
U2 - 10.1109/BHI.2019.8834624
DO - 10.1109/BHI.2019.8834624
M3 - Conference contribution
AN - SCOPUS:85073035877
T3 - 2019 IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2019 - Proceedings
BT - 2019 IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2019 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2019 IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2019
Y2 - 19 May 2019 through 22 May 2019
ER -