This paper introduces a new sentiment analysis model, BERT-LSTM-CNN (BLC), designed to classify students' reviews on Massive Open Online Courses (MOOCs). The study highlights the challenges of e-learning, particularly in MOOCs, where direct interaction between tutors and students is lacking, leading to low completion rates and student disengagement. Sentiment analysis is crucial for tutors to monitor student behavior and intervene when necessary. The proposed BLC model combines pre-trained BERT for word embedding, LSTM for semantic relationships, and CNN for capturing local features. The model outperforms existing machine learning and deep learning models in sentiment analysis, providing a robust tool for adaptive educational systems. The research aims to enhance the effectiveness of online learning by leveraging advanced natural language processing techniques.This paper introduces a new sentiment analysis model, BERT-LSTM-CNN (BLC), designed to classify students' reviews on Massive Open Online Courses (MOOCs). The study highlights the challenges of e-learning, particularly in MOOCs, where direct interaction between tutors and students is lacking, leading to low completion rates and student disengagement. Sentiment analysis is crucial for tutors to monitor student behavior and intervene when necessary. The proposed BLC model combines pre-trained BERT for word embedding, LSTM for semantic relationships, and CNN for capturing local features. The model outperforms existing machine learning and deep learning models in sentiment analysis, providing a robust tool for adaptive educational systems. The research aims to enhance the effectiveness of online learning by leveraging advanced natural language processing techniques.