Author:
Fenerich Amanda Trojan,Romanelli Egídio José,Catai Rodrigo Eduardo,Steiner Maria Teresinha Arns
Abstract
The aim of this study is to apply Machine Learning techniques for predicting and classifying the stress level of people commuting from home to work and also to evaluate the performance of prediction models using feature selection. The database was obtained through a structured questionnaire with 44 questions, applied to 196 people in the city of Curitiba, PR. The classification algorithms used were Support Vector Machine (SVM), Bayesian Networks (BN), and Logistic Regression (LR), comparatively. The results indicate that the classification of stress levels of new instances (people) as “high” or “low” can be performed using the LR technique (presenting the highest accuracy, 83.67%).
Publisher
Universidade Federal de Pernambuco
Reference23 articles.
1. Blum, A. L., & Langley, P. (1997). Selection of relevant features and examples in machine learning. Artificial intelligence, 97(1-2), 245-271.
2. Deng, Y., Wu, Z., Chu, C. H., & Yang, T. (2012, August). Evaluating feature selection for stress identification. In 2012 IEEE 13th International Conference on Information Reuse & Integration (IRI) (pp. 584-591). IEEE.
3. Fernandez, R., & Picard, R. W. (2003). Modeling drivers’ speech under stress. Speech communication, 40(1-2), 145-159.
4. Guyon, I., & Elisseeff, A. (2003). An introduction to variable and feature selection. Journal of machine learning research, 3(Mar), 1157-1182.
5. Hennessy, D. A. (2008). The impact of commuter stress on workplace aggression. Journal of Applied Social Psychology, 38(9), 2315-2335.