Design and implementation of a cascaded adaptive neuro-fuzzy inference system for cognitive and emotional stress level assessment based on electroencephalograms and self-reports


Gokongwei College of Engineering


Electronics And Communications Engg

Document Type

Conference Proceeding

Source Title

2014 International Conference on Humanoid, Nanotechnology, Information Technology, Communication and Control, Environment and Management, HNICEM 2014 - 7th HNICEM 2014 Joint with 6th International Symposium on Computational Intelligence and Intelligent Informatics, co-located with 10th ERDT Conference



Publication Date



Stress has been considered as one of the culprit in many diseases and other physical disorders. There are several methods on how to determine stress levels which are usually conducted by an experienced clinician. However, computer aided detection systems could be more objective and consistent in delivering results as basis of diagnosis and suggestions for relieving stress. In this paper, a cascaded adaptive neuro-fuzzy inference system (ANFIS) was proposed to assess the stress level in the cognitive and emotional aspect of an individual using EEG and self-reports. The two-stage ANFIS was able to predict the level of confusion, difficulty and frustration of the respondents with the task given to them. Using these factors, the system was also able to predict the level of stress that they had. Results show close proximity to the expected levels as described by the respondents through a system evaluation using the root mean square error and a parametric statistical test for significant difference.



Electrical and Electronics | Systems and Communications


Electroencephalography; Stress (Physiology); Entropy

Upload File


This document is currently not available here.