Students from different cultures possess varying levels of skills in learning, remembering, and understanding concepts. Some terms and their explanations may seem easy for one group of students but difficult for another. Therefore, delivering educational content that aligns with student’s learning capabilities is a challenging task based on cultural orientations. This study addresses the learning challenges by developing a Thesaurus Glossary E-learning (TGE) framework method. This study introduces the TGE method which is a multi-language tool with visual associations that adapts to students’ capabilities. It also examines cultural differences and native languages, particularly aiding Arab Native to visualize appropriate terms (thesaurus) and their explanations (glossary) based on students’ learning capabilities. TGE learns from students’ term selection behavior and displays terms at a simple or advanced level that matches their learning ability. Additionally, TGE demonstrated its effectiveness as an e-learning tool, accessible to all students anytime and anywhere. The study analyzed 314 records related to student performance, out of which 114 students were surveyed to evaluate the effectiveness of the TGE method. This work presents TGE as a novel e-learning tool designed to enhance conceptual thinking within the context of modern educational practices during the digital transformation. TGE is based on artificial intelligence algorithms and associative rules that simulate the human brain, establishing logical connections between related key terms and sketching associations among diverse facets of a situation. An experiment was conducted at a private university in the Sultanate of Oman to assess the effectiveness of the proposed TGE tool. TGE was integrated with selected subjects in information systems and used by the students as a resource for e-learning methods and materials. The results show that 85% of students who used TGE improved their performance by 19%. We believe this work could establish a new smart e-learning teaching method and attract modern and digital universities to enhance student learning outcomes linked with conceptual thinking.

Machine analysis of detection of the face is an active research topic in Human-Computer Interaction today. Most of the existing studies show that discovering the portion and scale of the face region is difficult due to significant illumination variation, noise and appearance variation in unconstrained scenarios. To overcome these problems, we present a method based on Extended Semi-Local Binary Patterns. For each frame, an aggregation of the pixel values over a neighborhood is considered and a local binary pattern is obtained. From these a binary code is obtained for each pixel and then histogram features is computed. Adaboost algorithm is used to learn and classify these discriminative features with the help of exemplar face and non-face signature of the images for detecting the location of face region in the frame. This Extended Semi Local Binary Pattern is sturdy to variations in illumination and noisy images. The developed methods are deployed on the real time YouTube video face databases and found to exhibit significant performance improvement owing to the novel features when compared to the existing techniques.