An Effective English Online user Learning Behaviour System using Random Forest and Extreme Gradient Boosting

For the past few decades, English language proficiency is essential as it is the most common and dominant language in business, education, science and technology which is used to communicate with people across the globe. An effective English language learning system is required for providing students with the knowledge and abilities to succeed in academic environment, as well as for building their self-assurance and competence in social contexts and for analyzing English online user learning behavior. Hence, this paper demonstrates English language learning system using Random Forest (RF) and Extreme Gradient Boosting (XGBoost). Initially, Students feedback is collected using the questionnaire model namely Unified Theory of Acceptance and Use of Technology 2 (UTAUT-2) and then data validation is imposed using Structural Equation Modeling (SEM) for structural analysis. Then the processed data is fed as input to Generative Pre-trained Transformer 2 (GPT-2) for training the model and those data is given to an effective integrated classification model named as RF-XGBoost. From the results, it clearly shows that proposed RF-XGBoost predicted higher accuracy of 98.95% than the conventional Naive Bayes model.