Modeling physiological effects of linalool and linalyl acetate-based odorants on stress levels of college faculty members

Date of Publication


Document Type

Master's Thesis

Degree Name

Master of Science in Computer Science


College of Computer Studies


Computer Science

Thesis Adviser

Merlin Teodosia C. Suarez

Defense Panel Chair

Rafael Cabredo

Defense Panel Member

Jocelyn Cu
Merlin Teodosia C. Suarez


Stress has emerged as a prevalent problem in todays society, including the academic setting. This is particularly evident in teachers, who have to deal with a heavy workload and pressure from peers and students. Such stress cannot be measured directly. However, it can be represented by using physiological indicators of stress, such as heart rate and respiration levels, which are quantifiable. Some stress coping methods involve taking prescriptive medicine, or listening to music to help reduce stress. However, aromatherapy is another viable solution in coping with this occupational stress because it is cheap, non-invasive, portable, easy to use, and can be used to target a large group of subjects.

Ten subjects participated as subjects in this study. Their ages ranged between 21 and 60 years old. Of the ten subjects, three of the subjects are faculty who are teaching, as well as performing administrative duties. The test consisted of a baseline measurement, a Stroop test as a simulated stressor, and a rest period. Biosignal measurements were taken continuously throughout the test. This was repeated without odorants, and with lavender and bergamot odorants to serve as aromatherapy. Odorants were chosen based on prior studies on affects of odorants on stress. Finally, machine learning algorithms implemented by Weka were applied to the data. This was used to create a classification model based on the data gathered from experiments. The output is a classification model that is able to detect if a person is stressed or not, based on his/her biometric signals. A dataset of the effects of certain odorants on the measured stress levels was also created from the study.

Use of the lavender and bergamot odorants has been shown to cause a reduction in physiological indicators of stress. This suggests that the odorants have stress-reducing effects. Both odorants performed fairly similarly, suggesting that the presence of a pleasant odorant is more important than the specific odorant used.

Using a general model approach, an average of 73.53% accuracy and Kappa statistic of 0.4706 was achieved. With a user-specific model approach, the k-Nearest neighbor algorithm was able to achieve an accuracy of 86.10% with Kappa statistic of 0.7220. Use of the multilayer perceptron algorithm resulted in a 85.55% accuracy rate and 0.7109 Kappa statistic, while the support vector machine algorithm had an accuracy of 89.01%% and 0.7802 Kappa statistic.

Abstract Format






Accession Number


Shelf Location

Archives, The Learning Commons, 12F Henry Sy Sr. Hall

Physical Description

1 computer optical disc ; 4 3/4 in.

This document is currently not available here.