Learners Behaviour Prediction and Analysis Model for Smart Learning Platform using Deep Learning Approach

Abstract

In the quickly changing field of instructional technology, intelligent educational systems are now essential for individualized and effective instruction. To forecast and understand learners’ actions in intelligent educational settings, this research suggests an analytical framework that makes use of deep learning techniques. By offering real-time information on user activities, the goal is to improve these platforms’ reactivity and flexibility. Using state-of-the-art deep learning designs, our technique examines large datasets that include interactions between users, interest trends, and efficiency measures. The proposed method classifies the e-learning based behaviour classification and then the e-learning performance prediction using CNN-LSTM. The suggested framework incorporates the temporal relationships and sequential patterns present in learners’ actions on the platform by fusing convolutional neural networks (CNNs) and long short-term memory networks (LSTMs). Furthermore, using multimedia information like simulations that are interactive and video lectures, convolutional neural networks (CNNs) are used to gather spatial data. The present study advances smart learning technology by providing a stable and expandable structure for behavior analysis and prediction in students. Through proactive customization of learning events, instructors, content producers, and platform developers can create a setting that is both enjoyable and effective for students. This is made possible by the knowledge gained from this approach.