دانلود رایگان مقاله انگلیسی استفاده از دفاتر هوشمند برای پیش بینی استرس شغلی - الزویر 2018

ABSTRACT

Occupational stress is increasingly present in our society. Usually, it is detected too late, resulting in physical and mental health problems for the worker, as well as economic losses for the companies due to the consequent absenteeism, presenteeism, reduced motivation or staff turnover. Therefore, the development of early stress detection systems that allow individuals to take timely action and prevent irreversible damage is required. To address this need, we investigate a method to analyze changes in physiological and behavioral patterns using unobtrusively and ubiquitously gathered smart office data. The goal of this paper is to build models that predict self-assessed stress and mental workload scores, as well as models that predict workload conditions based on physiological and behavior data. Regression models were built for the prediction of the self-reported stress and mental workload scores from data based on real office work settings. Similarly, classification models were employed to detect workload conditions and change in these conditions. Specific algorithms to deal with class-imbalance (SMOTEBoost and RUSBoost) were also tested. Results confirm the predictability of behavioral changes for stress and mental workload levels, as well as for change in workload conditions. Results also suggest that computer-use patterns together with body posture and movements are the best predictors for this purpose. Moreover, the importance of self-reported scores' standardization and the suitability of the NASA Task Load Index test for workload assessment is noticed. This work contributes significantly towards the development of an unobtrusive and ubiquitous early stress detection system in smart office environments, whose implementation in the industrial environment would make a great beneficial impact on workers’ health status and on the economy of companies.

5. Discussions and conclusion

In this paper, we analyzed the possibility of predicting workers' stress and workload levels, as well as changes in these conditions, by means of time-series statistics computed from unobtrusively collected physiological and behavioral data in a smart office environment. The research questions in hands are of great interest to today's society where stress is becoming increasingly present and harmful, but are also pertinent to the current state of the art in ambient intelligence and smart environments. Unobtrusive monitoring of peoples' behavior and physiology is already possible, but we yet need to associate these patterns to the disorder of interest. Moreover, it is still necessary to clarify and limit the use of the proposed system to avoid ethical and privacy issues before is implementation (Alberdi et al., 2015). Results show that the prediction of perceived stress and workload levels is possible using change and variability patterns of data collected unobtrusively from smart offices.

A regression analysis of the target scores from smart office data showed many statistically significant results, enforcing the hypothesis that this kind of collected data can actually predict the perceived stress and workload levels. The correlations found by this analysis vary from moderate to strong, depending on the nature of the objective label. NasaTLX scores, together with effort, mental effort and valence were the best-predicted scores, whereas self-reported stress and performance didn't show enough statistical significance to be considered predictable. In case of stress prediction, this is not surprising, as this label was acquired by means of a single-question visual analog scale, which unlike NasaTLX, RSME or VAS questionnaires, is not a questionnaire whose reliability has been verified and might be too subjective to be well capturing the real perceived stress levels of the users. Nonetheless, the analyses on the standardized scores improved the previous results, even demonstrating predictability for the self-reported stress and performance levels. This reasserts the fact that there is some inter-subject variability present on every score used for the study, but also suggests that controlling for this variability by means of standardization methods, can make their prediction possible.