Publication

Abstract : Attention in students remains a crucial cognitive function that plays a basic role in learning, academic performance, and overall well-being. With attention, people can shut out interruptions and unimportant knowledge while concentrating their cognitive pro- cesses on particular tasks, stimuli, or data. In this work, a classification of student’s attention using objective-based assessment is performed. Eye-tracking Data for Learner Assessment in Object-Oriented Programming is analysed which can test student’s performance and their attention level in the assessment. Different machine learning such as Random forest, KNN, SVM, Decision tree, Naive Bayes, AdaBoost, GradientBoosting, and deep learning models such as CNN, LSTM, Hybrid model, GoogleNet, and Alexnet are considered. To estimate the attention of students while attending an object-oriented assessment using eye movements, the study considers whether students are attending or not attending the multiple-choice questions. Participants we considered know object-oriented languages. The dataset consists of seven eye movement features that are Fixation count, Fixation duration, Saccade count, Saccade Duration, Saccade Amplitude, Blink count, and Scan path Length. To obtain extra information on the viewing be- havior, feature engineering is used to obtain derived features. A comparative analysis of different models that classify the students’ attention and their impact is depicted.