FAFatima Alshannaq

Fatima Alshannaq

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Fatima Alshannaq has a total of 24 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Speaking Two Languages to One Model: Translation, Code-Switching, and Response Quality in a Bilingual Arabic--English LLM Coding Study

    by Omar Alomeir, Abrar Wafa, Mohammad Asmat Ullah Khan, Elmoatez Billah Nagoudi, Saeed Abu Oleet, Fatima Alshannaq

  • Open Access

    Bayesian-optimized convex neural networks with embedded feature selection for high-dimensional anthropometric health prediction

    by Faten S. Alamri, Sunusi Bala Abdullahi, Noor Ayesha, Fatima Alshannaq, Tanzila Saba

    Background and Objectives: Anthropometric prediction and metabolic disorder screening require cost-effective solutions in resource-limited settings. We develop a convex-optimized neural network with embedded feature selection to establish quantitative relationships between body morphology and metabolic health. Methods: Our framework combines Bayesian hyperparameter tuning with $\ell$1–$\ell$2 regularized feature selection (λ1 = 0.020, λ2 = 0.002), processing 4,464 3D body scans, 1,000 Iraqi clinical records, and 768 Pima Indian diabetes cases. Spatial fully connected networks preserve anatomical relationships during feature selection. Results: The model achieved 99.000% accuracy (95% CI [98.750–99.250]) for anthropometric gender classification, 94.700% accuracy (95% CI [93.200–96.200]) for Iraqi diabetes triage, and 90.040% accuracy (95% CI [86.800–91.280], Area Under the Receiver Operating Characteristic Curve (AUC) = 0.941) for Pima diabetes prediction, outperforming conventional methods (paired t = 27.340, p < 0.001). Key findings include identification of spine-to-scye length (weight = 0.990, Pearson r = 0.920) and Very Low-Density Lipoprotein (VLDL) (weight = 0.990, r = 0.982) as previously unreported predictors requiring further investigation, alongside traditional indicators. Conclusion: The framework demonstrates anthropometry’s viability for security and diagnostic screening, with 57% parameter reduction vs . ensemble baselines while maintaining application-grade performance.

  • Leveraging Deep Learning for Cyberbullying Detection in Arabic Tweets

    by Fatima AlShannaq, Sawsan Alhalawani, Heba Al-Jarrah, Mohammad Shehab

    Cyberbullying is a growing concern in the digital age, particularly on social media platforms like X (formerly Twitter). This study presents a deep learning-based approach for detecting cyberbullying in Arabic tweets using the Arabic Cyberbullying Corpus (ArCybC). Leveraging the Arabic-BERT model, we conducted binary and multi-class classification tasks to identify cyberbullying content and its associated categories or domains. Our methodology incorporated advanced data-balancing techniques, including oversampling and undersam-pling, to address class imbalance challenges. The study aims to answer the research questions: (1) How effectively can deep learning models detect cyberbullying in Arabic tweets? (2) How do data balancing techniques impact classification performance? (3) Can Arabic-BERT outperform traditional classifiers in cy-berbullying detection? These questions are addressed through our methodology, where we employ Arabic-BERT, implement rebalancing techniques, and compare our model's performance with baseline classifiers. Experimental results demonstrate significant improvements in F1-scores, with 89.1% achieved in binary classification and 91.6% in category-level classification after oversampling. These findings highlight the effectiveness of fine-tuned deep learning models in tackling the linguistic complexity of Arabic and the nuanced nature of cyberbullying detection. This study provides a foundation for future research in addressing online harassment in Arabic-speaking digital communities.


  • An Exploration Into the Mechanisms and Evolution of GPT Models

    by Fatima Barakat AlShannaq, Mohammad M. Shehab, Anwar H. Al-Assaf, Esra'A Alhenawi, Shatha Awawdeh

    This chapter of the book explores the development and capabilities of Generative Pre-trained Transformers (GPT) models, tracing their evolution and comparing them with other AI language models. It delves into practical applications across domains, emphasizing GPT's role in generating human-like text, aiding natural language understanding, and supporting conversational AI. The chapter discusses the profound implications of GPT's capabilities, including opportunities for automating content creation. Additionally, it critically examines limitations and ethical concerns, addressing issues like biased language generation, potential misinformation, and privacy and data security. The overall aim is to provide a comprehensive and responsible perspective on developing and deploying GPT models.


  • Open Access

    Offensive Language Detection in Arabic Social Networks Using Evolutionary-Based Classifiers Learned From Fine-Tuned Embeddings

    by Fatima Shannaq, Bassam Hammo, Hossam Faris, Pedro A. Castillo-Valdivieso

    Social networks facilitate communication between people from all over the world. Unfortunately, the excessive use of social networks leads to the rise of antisocial behaviors such as the spread of online offensive language, cyberbullying (CB), and hate speech (HS). Therefore, abusive\offensive and hate detection become a crucial part of cyberharassment. Manual detection of cyberharassment is cumbersome, slow, and not even feasible in rapidly growing data. In this study, we addressed the challenges of automatic detection of the offensive tweets in the Arabic language. The main contribution of this study is to design and implement an intelligent prediction system encompassing a two-stage optimization approach to identify and classify the offensive from the non-offensive text. In the first stage, the proposed approach fine-tuned the pre-trained word embedding models by training them for several epochs on the training dataset. The embeddings of the vocabularies in the new dataset are trained and added to the old embeddings. While in the second stage, it employed a hybrid approach of two classifiers, namely XGBoost and SVM, and a genetic algorithm (GA) to mitigate the drawback of the classifiers in finding the optimal hyperparameter values to run the proposed approach. We tested the proposed approach on Arabic Cyberbullying Corpus (ArCybC), which contains tweets collected from four Twitter domains: gaming, sports, news, and celebrities. The ArCybC dataset has four categories: sexual, racial, intelligence, and appearance. The proposed approach produced superior results, in which the SVM algorithm with the Aravec SkipGram word embedding model achieved an accuracy rate of 88.2% and an F1-score rate of 87.8%.