Artificial Intelligence
Mujtaba Sultani; Negin Daneshpour
Abstract
Background and Objectives: Discussion forums in Massive Open Online Courses (MOOCs) enable students to interact with instructors and share educational concerns. However, identifying urgent posts within the vast volume of discussions poses significant challenges. High dropout rates and the need for timely ...
Read More
Background and Objectives: Discussion forums in Massive Open Online Courses (MOOCs) enable students to interact with instructors and share educational concerns. However, identifying urgent posts within the vast volume of discussions poses significant challenges. High dropout rates and the need for timely responses to urgent queries highlight the importance of efficient classification systems. This study addresses the binary classification of student posts in the Stanford MOOC Posts dataset into urgent and non-urgent categories, and aims to improve performance in the presence of class imbalance.Methods: We propose a hybrid deep learning model that integrates BERT-based contextual embeddings with a Convolutional Neural Networks (CNN) and Bidirectional Long Short-Term Memory (BiLSTM) architecture to capture both local textual features and long-term dependencies. To mitigate the class imbalance issue, BERT-based data augmentation technique was employed which enriches minority class samples, and enhance model generalization and urgent post detection. The model’s performance was compared against baseline methods, including CNN, LSTM, BiLSTM, and other state-of-the-art models. Evaluation metrics such as F1-weighted score and class-specific F1 scores were used to assess effectiveness.Results: The model achieved a 93.3% F1-weighted score and an 84.1% F1 score for the urgent class which surpasses the best-performing state-of-the-art model by 0.6% and 0.8%, respectively. The results show the effectiveness of augmentation and hybrid architecture while identifying urgent posts, particularly within imbalanced datasets.Conclusion: This research introduces a scalable and effective framework for urgent post detection in MOOCs. By leveraging BERT-based augmentation and a CNN–BiLSTM hybrid architecture that integrates contextual and sequential learning, the study provides automated forum analysis, offer timely insights for instructors and course designers to enhance students support, engagement, and retention.
Classification
Mujtaba Sultani; Negin Daneshpour
Abstract
Background and Objectives: Massive Open Online Courses (MOOCs) face unique challenges in extracting student feedback from large, asynchronous student discussion forums. While traditional survey methods are commonly used, they struggle with scalability and real-time analysis in the MOOC context. This ...
Read More
Background and Objectives: Massive Open Online Courses (MOOCs) face unique challenges in extracting student feedback from large, asynchronous student discussion forums. While traditional survey methods are commonly used, they struggle with scalability and real-time analysis in the MOOC context. This study aims to address these limitations and focus on automated extraction and classification of student opinions and their urgency. The study bridges the gap between suggestion mining in commercial applications and educational domains.Methods: We presented a novel deep learning approach using a BERT-based hybrid Convolutional Neural Network (CNN) – Bidirectional Long Short-term Memory (BiLSTM) multi-output model, named CBiLSTM. The model was trained to classify student posts into opinions and further categorize them by urgency. Performance metrics such as F1-weighted scores, Precision-Recall curves, and Area Under the Curve (AUC) were used to evaluate the model's efficacy, particularly in handling imbalanced datasets.Results: The presented CBiLSTM model got F1-weighted score of 87.3% for opinion classification and 81.1% for urgency classification which represents an improvement of 1.3% and 1.8% over the best-performing baseline model. Precision-Recall curves and AUC metrics highlights the model's strength in balancing precision and recall. These findings demonstrate the model's capacity to accurately classify and prioritize student feedback in the educational domain.Conclusion: This study offers a robust framework to enhance decision-making processes in MOOCs through effective feedback analysis. The CBiLSTM model provides a scalable, data-driven solution that empowers instructors, course designers, and policymakers to make targeted improvements, and improves student engagement and course quality.
