Improving Academic Performance and Career Mobility Through Hybrid Clustered Graph Neural Networks

ABSTRACT The main concern of the intelligence course recommendations is to improve college students' innovation and entrepreneurship learning experience. Thus, the need for individualized effective materials in modern education increases as much as the rates of online education platforms. Moreover, this expansion usually comes with various related drawbacks, and one of them is the problem of searching for classes that meet the learners' preferences and goals. When it comes to educational data, traditional methods of data processing fail to control such a huge amount of data and might even lead to distortions. To this end, this study presents the Hybrid Clustered Graph Neural Network to provide a more accurate analysis and prediction of students' academic performance for providing course recommendations. An efficient course recommendation framework named Hybrid Clustered Graph Neural Network is proposed for the career development of engineering students. The descriptor datasets were used for this research article which contains the details of course and user requirements. The collected descriptor data are preprocessed by imputation and normalization approaches to provide the enhanced quality and relevance of the data. In the feature extraction phase, the Clustering‐based Graph Convolutional Representation model is implemented to extract student's recommendations and WordPieceFormer is applied for the extraction of contextual‐based social media features. The Hybrid Clustered Recurrent Neural Network model is proposed for scoring and ranking the courses according to the recommendation ranking aspects. This study examines the behavioral performance using the proposed approach, providing appropriate course suggestions to achieve career mobility objectives. The evaluations indicated the viability of the proposed model, showing an accuracy efficiency of 98% and precision of 96.6%. The following results show the benefits of the proposed approach in attaining the appropriate recommendations that meet the students' academic performance and student career needs for providing course recommendations.