Twitter is the most popular micro-blogging medium that allows users to exchange short messages, provides a platform for public people to share the news. Nowadays, Twitter counts with an average of 328 million monthly active users and is growing rapidly. Detecting the credibility of shared information on Twitter becomes a necessity, especially during high impact events. In this paper a classification model based on supervised machine learning techniques is proposed to detect credibility. The proposed model uses an extensive set of features including both content-based and source-based features. The research compares the performance of five different machine learning classifiers using three feature sets: content based, source based and a combination of both sets. The best performance is achieved when using a combined set of features and applying Random Forests as a classifier with accuracy 78.4%, precision 79.6%, recall 91.6% and f1-measure 85.2%. Experiments also revealed that the proposed model achieves improvement of 22% when compared to CRF which applies the same approach in terms of F1-measure. Feature analysis is presented to highlight the importance of the source-based features compared with the content-based features as deciders for credibility.

Automated scoring systems have been revolutionized by natural language processing, enabling the evaluation of students’ diverse answers across various academic disciplines. However, this presents a challenge as students’ responses may vary significantly in terms of length, structure, and content. To tackle this challenge, this research introduces a novel automated model for short answer grading. The proposed model uses pretrained “transformer” models, specifically T5, in conjunction with a BI-LSTM architecture which is effective in processing sequential data by considering the past and future context. This research evaluated several preprocessing techniques and different hyperparameters to identify the most efficient architecture. Experiments were conducted using a standard benchmark dataset named the North Texas Dataset. This research achieved a state-of-the-art correlation value of 92.5 percent. The proposed model’s accuracy has significant implications for education as it has the potential to save educators considerable time and effort, while providing a reliable and fair evaluation for students, ultimately leading to improved learning outcomes.

Natural language processing (NLP) has become somewhat well-known because of its many uses; deep neural networks have driven major developments. Still, there are difficulties, especially in Arabic NLP, where the language’s large vocabulary of over 12 million words and several dialects cause special issues. Arabic has a large speaker base; however, NLP studies in this language find challenges, particularly with class imbalance. Many times, standard class balancing methods overlook intra-class similarity, a crucial element influencing model training. We present a new approach for computing intra-class similarity using cosine similarity and embedding models to find ideal class weights for model training, hence bridging this difference. On two benchmark datasets—the Arabic Semantic Question Similarity dataset (NSURL) and the Microsoft Research Paragraph Corpus (MRPC)—we assessed our method. With an accuracy of state-of-the-art 83.25% on the MRPC dataset and 96.931% on the NSURL dataset, our approach proved successful in improving model performance in Arabic text classification.

This paper explores the potential of Large Language Models (LLMs) in automating the grading and feedback generation of short-answer responses in Arabic. Arabic poses unique challenges due to its linguistic complexity and the relative scarcity of well-developed Natural Language Processing (NLP) resources compared to languages such as English and Chinese. The study evaluates both proprietary models (GPT-4) and open-source models (Llama 3-8B, Llama 3-70B, and DeepSeek-V3) using the Environmental Science Corpus—a custom-designed dataset tailored for Arabic short-answer assessment. Two core tasks are addressed: grading and feedback generation. In the grading task, DeepSeek-V3 achieved the best performance, with a Quadratic Weighted Kappa (QWK) score of 0.8273, a Pearson correlation of 86.09%, and a Root Mean Squared Error (RMSE) of 0.76, indicating near-perfect agreement with human evaluators. GPT-4 ranked second, followed by Llama 3-70B, while Llama 3-8B was the lowest-performing model. In feedback generation, DeepSeek-V3 again led the performance with a human evaluation score of 79.61% for generating accurate and constructive feedback. GPT-4 ranked second, followed by Llama 3 models. Statistical analysis using the Wilcoxon test revealed significant performance differences among all models (p < 0.05), indicating that each LLM offers unique capabilities in handling Arabic short-answer grading. Overall, the results underscore the effectiveness of LLMs in Arabic educational assessment, highlighting the critical role of prompt engineering in enhancing model performance. The study demonstrates that LLMs can not only grade student responses with high accuracy but also generate meaningful feedback, thereby supporting the development of more effective automated learning tools. Practical recommendations and best practices are presented to help educators and developers optimize the use of LLMs in Arabic-language educational settings, laying the groundwork for future advancements in Arabic NLP.