Deep Flamingo Search and Reinforcement Learning Based Recommendation System for E-Learning Platform using Social Media

Abstract

The e-learning platform offers ideal courses for learners depending on their skills and knowledge. Nowadays, scientific researchers and educational centers are trying to improve online education by finding new approaches and optimal methods. Nowadays, different kinds of recommender systems have been used to select proper courses and optimal paths for learners, but these systems can suffer from many problems. So this research introduces a novel recommendation system for learners for choosing relevant courses with the help of social media. The proposed methodology is based on the clustering process, the creation of learner profiles and deep flamingo search reinforcement learning (DFSRL) based recommendation system. The two parameters, productivity and motivation, are used for the clustering process. A semantic similarity-based Fuzzy logic method is used to classify sentiments into positive, negative or neutral classes. Then, create a learner's profile based on the ratings from the quiz result. Also, our proposed system presents a new reinforcement learning method known as the Deep Flamingo Search reinforcement learning-based recommendation system to discover the best path the learner must follow and select the proper course. The proposed method is implemented on Python software, and its performance is analyzed for accuracy, precision, recall, F1-score.