الفهرس 
Hybrid approach to recognize learners behaviors using facial expressions and machine learning
AbstractOnly 14 pages are availabe for public view
 |  | from | 129 |  |  |
| | | from | 129 | | |
Abstract
Computer based learning environments are technology based instructional tools used for educational purpose. Usually, these environments take into consideration the learner’s mental responses that are based on solving tests and exams questions to determine the next level of the learning process and manage the learning flow of the learner.They however suffer from engaging the emotional behaviors and facial expressions of the learner that reflect the learner’s emotional states during the learning process.The later engagement could make the learning flow more adaptive as each learner could be redirected to a learning flow that matches his capabilities and academic performance.This thesis proposes a novel approach for modeling an adaptive computer based learning environment by considering the integration between the learner mental responses to exams questions and his emotional states during the exam sessions. In the proposed approach, a convolutional neural network is used to detect and analyze the learners’ facial expressions, and outperforms other CNN models on the same training benchmark.The fuzzy system is used to determine the next learning level based on several response and interaction factors of the learner including emotional interaction factors and mental responses factors. The thesis also introduces corpora for evaluating the performance of the proposed approach