IRJMST Vol 7 Issue 12 [Year 2016] ISSN 2250 – 1959 (0nline) 2348 – 9367 (Print) International Research Journal of Management Science & Technology http://www.irjmst.com Page 151 Modelling Characteristics of Eye Movement Analysis for Stress Recognition using Artificial Neural Network M. S. Kalas email:mam2kalas@rediff.com Research scholor,Walchand College of Engg,Sangli B. F. Momin email:bfmomin@yahoo.com Asso. Prof,Dept of CSE, Walchand College of Engg,Sangli Abstract The principal reason for measuring stress is to quantify the mental cost of performing tasks in order to predict operator and system performance. We have to characterize mental states of operator performance, by finding patterns in timely changing physiological, measures like EOG(Electro Occulography), with eye blinks. Various computational approaches based on EOG signals have been developed for analyzing and detecting stress of an individual. We report Artificial Neural Network (ANN) architecture that unfolds innovation in Neuroimaging as an approach to handle EOG data analysis by means of classifying Ground truth annotation. This exploration exhibits optimum ANN architecture for the modelling of EOG data by varying the network properties such as the network type, training function, transfer function and number of neurons in hidden layer. ANN architecture hence inferred consists of nonlinear sigmoid activation function for hidden layer and Levenberg-Marquardt backpropagation method for training the model. We have done the comparative analysis of eye movement analysis for stress detection.We validate the method using an eight participant study in an office environment using an example set of two activities,Physical activity related to operating heavy machines, ,Cognitive activities like working with computers: copying a text, reading a printed paper, taking handwritten notes, watching a video, and browsing the Web. We also include periods with no specific activity (the NULL class). The result suggests that the ANN has the potential to exhibit as the best tool for modeling of protein samples, thus learning by examples can be achieved. Keywords: ElectroOcculography, ANN architecture ,stress detection. 1. INTRODUCTION Stress detection is an ongoing research topic among both psychologists and engineers. Various technologies are developed on human stress detection using wearable sensors and bio signal processing. ElectroOcculography (EOG) data is used to develop and adjust the method of stress detection in operators and academicians who are engaged in continuous reading and learning operations. By monitoring the eyes, and EOG analysis it is believed that the symptoms of operator’s stress can be detected early enough to avoid mistakes. There are several human activities where the awareness and conscious control is a very important factor: vehicle driving, heavy equipment operation, hazardous materials manipulation[4][5].