Learner Attention Detection and Dynamic Tutorial Recommendation using Image Processing Techniques

Computer Science and Engineering

India is always considered as one of the largest network of educational institutions. Although several constraints are been associated with our learning system. We try to provide the same content of teaching to all students with different inter personal skills. The most important factor is lack of student motivation towards a subject, course etc. Adaptive learning is an academic method that utilizes computers as an interactive teaching device. In most of the educational models, engagement is not monitored explicitly, but rather they assume the engagement based on the student’s responses to questions and tasks and provide the learning based on it. Thus dynamic student behavior analysis is a first step towards an automated teacher feedback tool for measuring student engagement. In our system, we propose a hybrid architecture system invoking student facial emotion recognition, eye gaze monitoring, head movements identifications based analyzing dynamic student engagement / behavior in classroom and towards a specific course at e-learning platforms. Our proposed architecture uses feature extraction algorithms like Principal Component Analysis (PCA) for facial emotion recognition, Haar Cascade for pupil detection and Local Binary Patterns for recognizing head movements. For machine learning approach and to provide accurate results we propose Open CV. Thus based on the students input weightage is allocated, based on the final score, we do compare with the threshold value. If the students attention value is greater than the threshold value, theory based deliverance is recommended. If the students attention value is lesser than the threshold value, video, smart class, motivational video based deliverance is recommended. For Experimental results are been implemented using Pycharm tool.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.