Saudi Cultural Missions Theses & Dissertations
Permanent URI for this communityhttps://drepo.sdl.edu.sa/handle/20.500.14154/10
Browse
113 results
Search Results
Item Restricted AI-Enabled Autonomous Knowledge Extraction from Large-Scale Textual Data(Saudi Digital Library, 2026) Alharbi, Abdulrahman; Obradovic, ZoranThe rapid growth of large-scale textual data across social media platforms, news media, and scientific repositories presents both unprecedented opportunities and significant challenges for extracting meaningful insights. During global events such as the COVID-19 pandemic, understanding public discourse requires analyzing vast amounts of noisy, heterogeneous, dynamic, and geographically distributed data. At the same time, the exponential increase in scientific publications has made traditional evidence synthesis methods increasingly labor-intensive, time-consuming, and difficult to scale. Existing approaches to textual knowledge extraction often operate in isolation, lack interpretability, fail to integrate heterogeneous data sources, and do not support scalable end-to-end automation. This dissertation addresses these limitations by proposing a unified framework for AI-enabled autonomous knowledge extraction from large-scale textual data. The research introduces a comprehensive pipeline that integrate sentiment analysis, topic modeling, semantic interpretation, spatiotemporal reasoning, and multi-agent automation for scalable, robust and reproducible text analysis across heterogeneous domains. First, this work introduces TriLex, a novel unsupervised sentiment analysis framework that combines multiple lexicon-based sentiment analysis methods through weighted aggregation, majority voting, and dynamic thresholding technique to improve robustness and accuracy for short and noisy textual data. Building on this foundation, a hierarchical spatiotemporal framework is developed to capture the evolution of public sentiment across global, national, and regional scales. The framework integrates over 7 million social media posts and thousands of news articles to analyze COVID-19 vaccine discourse across time, geographic regions, and platforms. To enhance topic interpretability, this research integrates BERTopic with large language models (LLMs), enabling automated generation of coherent and context-aware topic representations for large-scale textual discourse. A cross-platform analytical framework is further introduced to examine temporal relationships between social media and news media discourse, demonstrating a bidirectional relationship in which news coverage and public discourse influence each other over time. Extending beyond discourse analysis, this dissertation introduces an Agentic AI framework that automates the end-to-end process of large-scale multilingual knowledge extraction and evidence synthesis. The proposed multi-agent system coordinates specialized agents for query generation, multilingual retrieval, metadata harmonization, title and abstract screening, and full-text analysis. Evaluated on a multilingual corpus of over 52,000 scientific records, the framework achieves high screening performance while substantially reducing processing time from months to hours, demonstrating significant improvements in scalability, robustness, and reproducibility. Collectively, this dissertation bridges the gap between analytical understanding and autonomous knowledge extraction from large-scale textual data. By integrating robust sentiment analysis, interpretable topic modeling, spatiotemporal discourse analysis, and autonomous AI systems within a unified framework, this work establishes a scalable and extensible paradigm for transforming heterogeneous textual data into actionable knowledge. The proposed methodologies are validated using real-world datasets spanning social media, news media, and scientific literature across diverse application domains.13 0Item Restricted Artificial intelligence for bias detection in higher education online content(Saudi Digital Library, 2026) Bin Shiha, Rawan; Eric, Atwell; Noorhan, AbbasThis thesis develops and evaluates Artificial Intelligence (AI) and Natural Language Processing (NLP) approaches for detecting bias in higher education online resources, addressing the lack of systematic computational methods in this area. Bias in higher education content shapes knowledge production, representation, and equity, making its detection both academically and socially significant. To address this gap, three sets of novel datasets were created: 1. a corpus of university news articles annotated for subjectivity, sentiment, and gender representation; 2. three university reading list datasets with demographic annotations enabling comparative analysis across Western and Middle Eastern contexts; and 3. two domain-specific collections of learning materials, one from humanities-oriented open resources and the other from Science, Technology, Engineering, and Mathematics (STEM) lecture transcripts. Together, these datasets provide the first systematic resources for investigating representational, stereotypical and linguistic bias across diverse higher education domains. Using these resources, a range of Pre-trained Language Models (PLMs) and Large Language Models (LLMs) were evaluated for bias detection. PLM revealed significant gendered and representational disparities in university discourse, while fine-tuned LLM achieved improved performance on humanities data but showed limited transferability to STEM materials. A hybrid framework integrating fine-tuned LLMs with Retrieval-Augmented Generation (RAG) enhanced detection transparency and prediction balance. These methods were operationalised in a prototype web application for bias detection in higher education learning content. The contributions of this research are fourfold: 1. the creation of multiple novel annotated datasets spanning university news, reading lists, and academic learning resources; 2. the introduction of LLM-based strategies for bias annotation, fine-tuning, and cross-domain evaluation 3. methodological innovations including a structured framework for categorising bias, a hybrid human–AI annotation approach, and a replicable NLP pipeline for demographic and thematic analysis; and 4. the design of a hybrid bias detection system and accompanying web application. This work advances computational approaches to bias detection, provides reproducible resources for future research, and offers practical tools to support greater fairness and equity in higher education online environments.3 0Item Restricted Artificial Intelligence in Routine Non-contrast CT Imaging to Assess Cardiothoracic Structures and Evaluate Clinical Utility(Saudi Digital Library, 2026) Alnasser, Turki; Swift, Andrew; Alabed, SamerBackground: Pulmonary hypertension (PH) is a progressive and life-threatening condition characterised by elevated pulmonary arterial pressure and associated with different diseases, including left heart and lung diseases. Early diagnosis is essential to improve clinical outcomes; however, current diagnostic pathways rely on invasive right heart catheterisation or contrast-enhanced imaging, which are not always feasible in routine clinical practice and are associated with different complications. Non-contrast chest computed tomography (CT) is widely available and frequently performed in patients with PH, many of whom also exhibit coronary artery calcification. However, its full diagnostic and prognostic potentials remain underexplored. Although several manually derived CT measurements have been proposed as imaging predictors, their clinical utility is limited by inter-observer variability and time-consuming analysis. Aim: This thesis investigates the use of artificial intelligence (AI)–driven volumetric measurements of multi-cardiothoracic structures from non-gated, non-contrast CT to enhance both the diagnosis and prognostic assessment of PH and coronary artery calcification. Methods: A multi-cardiothoracic structure and coronary artery calcification AI-based segmentation models were developed at the University of Sheffield using internal and external cohorts from the ASPIRE registry. The models were benchmarked against the gold standard haemodynamic, reference standards, visual assessments, and validated tool (e.g. TotalSegmentator). Results : The developed AI-based models demonstrate high diagnostic accuracy in predicting PH and detecting coronary artery calcifications. The models were comparable to TotalSegmentator and demonstrated higher accuracy than manual clinical practice techniques when evaluated in exploratory testing. AI-derived right atrial volume and coronary artery calcifications are independently associated with increased mortality in PH patients, even after adjustment for age, sex, PH subgroups, and REVEAL score. Conclusion: Automated segmentation and volumetric measurements of multi-cardiothoracic structures in non-contrast CT have the potential to facilitate earlier and accurate diagnosis and prognostic assessment of coronary artery calcification and PH patients, including lung and left heart diseases.18 0Item Restricted Exploring the Utilization of Multimodal LLMs on Personalized Learning in Higher Education(Saudi Digital Library, 2025) Aljumaah, Jana Abdullah; Rana, Muhammad; Rizwan, TaimoorUse of Large Language Models (LLM) is rising in popularity amongst students in higher education. It has transformed the way students tackle their academic work, offering stu- dents powerful new tools for explanation, idea generation, writing support, and independent study. While their adoption has been widely discussed in theory, there remains limited empirical evidence on how students are using these technologies, how they per- ceive their benefits and risks, and how such practices may reshape learning. This dissertation explores the perceptions and practices of 55 UK university students re- garding the use of LLMs in academic work. Using a survey-based methodology, the study examines five key objectives: to identify patterns of LLM use, analyze the purposes for which they are employed, evaluate student attitudes toward their advantages and limita- tions, assess their influence on study habits and independent learning, and provide insights for ethical and effective integration into higher education. The results reveals that students use LLMs as tools for understanding and organizing knowledge such as summarization or concept explanations and clarifications. Secondly, they employ them for supporting their academic writing and production like proofreading, editing, or as a writing assistant. Likert-scale analyses indicate that students generally per- ceive LLMs as both useful and easy to use, consistent with the Technology Acceptance Model (TAM). However, concerns around trust and over-reliance temper this enthusiasm, highlighting a critical awareness of the risks involved. The findings suggest that while LLMs are integrated into study habits and promote efficiency, their long-term value will depend on guidance, ethical frameworks, and digital literacy training. The study contributes to the growing body of research on AI in education by offering evi- dence from the UK universities context. It demonstrates how perceptions of usefulness, ease of use, and reliability shape adoption, and it paves the way for future research, includ- ing cross-cultural comparisons, to better understand the evolving role of LLMs in higher education.40 0Item Restricted Efficient Language Model Compression For Mobile Deployment : A Study on Qwen1.5-1.8b-chat and Phi-3-Mini-4K-Instruct(Saudi Digital Library, 2026) Albalawi, Atheer Mohammed; Aletras, NikosLarge language models (LLMs) offer impressive capabilities but are often too computationally demanding for deployment on resource-constrained devices such as mobile platforms. This project investigates the effectiveness and limitations of compression techniques applied to two representative models, Qwen1.5-1.8B-Chat and Phi-3-Mini-4K-Instruct, using different approaches for each architecture. For Qwen1.5-1.8B-Chat, structured pruning and INT8 quantization were applied, achieving significant size reduction with high benchmark retention (88.4% MMLU, 95.9% ARC-Challenge). However, functional evaluation revealed severe degradation, including deterministic output collapse and loss of generative ability, indicating gaps in benchmark-driven evaluation. In contrast, Phi-3-Mini-4K-Instruct was compressed using EntroLLM mixed quantization and entropy-based methods, which preserved both accuracy and generative behavior, demonstrating greater deployment reliability. These findings highlight that compression outcomes are highly model-dependent and that standard benchmarks may obscure critical failures. This work contributes technical insights by identifying architecture-specific vulnerabilities to compression as well as methodological lessons that underscore the need for comprehensive, deployment-aware evaluation frameworks to ensure reliable LLM performance in practice.10 0Item Restricted Kem Boi – AI Content Generation System(Saudi Digital Library, 2026) Ghadeer Waleed Alkhunaizi, Chen-Ling Lin، Ranaweera Mudiyanselage Kushlani Dinuthri Kumari Amunugama and Saketh Reddy Jambula; Chen, DavidThe KemBoi AI Content Generation System was developed as part of a Griffith University Work Integrated Learning (WIL) project in collaboration with ApyHome Pty Ltd, a Gold Coast-based dessert brand. The project aimed to design and implement an AI-driven platform capable of automating the generation of marketing content, including ideas, scripts, and promotional video outputs, addressing the time-consuming and resource-intensive nature of traditional content creation processes. The proposed solution introduces a structured AI-powered content generation pipeline that transforms user prompts into complete marketing assets through automated stages, including idea generation, script creation, prompt processing, cinematic video generation, and final media assembly. The project adopted the Agile Scrum methodology and was implemented using HTML, CSS, JavaScript, Python, Flask, SQLite, and external AI APIs such as Together AI. Key deliverables included a web-based dashboard for job creation and monitoring, AI-powered modules for script and video generation, structured database storage, content management capabilities, and job-tracking mechanisms. Security and validation controls, including API key protection and input validation, were also incorporated into the system design.24 0Item Restricted Do Generative Chatbot Persuade Users? An Elaboration Likelihood Model Investigation with Cognitive and Emotional Trust(Saudi Digital Library, 2026) Mohanna, Sohaib; AlSurmi, AbdulrahmanGenerative chatbots are increasingly used as consumer information sources, yet their effects on how consumers process and adopt information remain insufficiently understood. It is unclear how the persuasive communication cues used by these chatbots influence a user’s perception of information usefulness and the likelihood of adopting that information. Existing studies in generative chatbots persuasion and information adoption have rarely integrated their theoretical perspectives in this context, focusing mostly on technology acceptance rather than information processing, leaving a gap in understanding how persuasive processes influence in conversations with such systems. Addressing this gap, the study applies an extended Information Adoption Model (IAM), a framework rooted in a dual process theory of persuasion, specifically the Elaboration Likelihood Model (ELM), to generative chatbots dialogue. It investigates how specific persuasive cues in generative chatbots influence users' cognitive and emotional trust, subsequently affecting their perceived information usefulness and adoption decisions. By extending the IAM with a multidimensional view of trust mechanisms (cognitive and emotional), this research contributes to the persuasion process of artificially intelligent human interactions by demonstrating how persuasive mechanisms participate into perceived information usefulness, the critical mediator preceding information adoption in generative chatbots contexts. To investigate these relationships, this research employs a positivist quantitative, theory-testing design. Data from 438 experienced chatbot users were collected, processed for analysis and the proposed model was tested using PLS-SEM. Analysis of the survey data supports the integrated model and reveals that both central and peripheral cues play significant roles via distinct pathways. Central cues, such as high information quality and relevance in chatbots responses, significantly impact cognitive trust and directly enhance consumers’ perceptions of information usefulness. Peripheral cues, including the chatbots human-like characteristics, primarily influence emotional trust. In turn, both cognitive and emotional trust emerge as strong positive predictors of perceived information usefulness and of consumers’ willingness to adopt chatbots information. Interestingly, opposite to the ELM contention of central dominance over peripheral cues, emotional trust exerted stronger influence toward information usefulness. This study makes both theoretical and practical contributions. Theoretically, it demonstrates the value of bridging persuasion and information adoption theories in the field of human–AI communication. By presenting how cognitive and peripheral cues respectively foster cognitive versus emotional trust, the research provides a nuanced understanding of how cognitive and superficial cues through which -provided information becomes persuasive and valuable to consumers. It also extends information adoption models by introducing a multidimensional view of trust perspective and specific chatbots constructs to highlighting that rational credibility and emotional rapport are both critical for consumers’ acceptance of information. In practical terms, the findings offer guidance for designing more effective generative chatbot systems.19 0Item Restricted ENHANCING TRAFFIC SAFETY THROUGH AI-DRIVEN, PRIVACY-PRESERVING, AND SECURE IMPAIRED DRIVING DETECTION SYSTEMS(Saudi Digital Library, 2026) Alsulieman, Razan; Sherif, AhmedDrunk driving remains a major threat to road safety worldwide, contributing significantly to traffic injuries and fatalities each year. Traditional detection approaches are largely reactive and vehicle-centric, relying on in-vehicle sensors, breathalyzers, or post-incident enforcement. These methods often depend on driver cooperation, intrusive hardware installations, or limited monitoring environments, restricting their scalability and effectiveness in large transportation systems. At the same time, modern cities increasingly deploy roadside cameras, surveillance networks, and drone based monitoring systems, creating new opportunities for proactive intoxication detection at the infrastructure level. However, leveraging such external monitoring introduces challenges related to secure data collection, reliable AI-based analysis, privacy protection, and real-world deployment. This dissertation proposes a secure, privacy-preserving Artificial Intelligence framework for proactive drunk driving detection using out-of-vehicle surveillance data. The framework addresses three key aspects required for reliable infrastructure-level monitoring. First, a lightweight authentication scheme is developed to ensure secure data collection from distributed monitoring platforms such as drones and surveillance devices. The proposed design employs physically unclonable functions and symmetric cryptographic primitives to provide protection against impersonation, replay attacks, and device cloning while maintaining low computational overhead for resource-constrainedenvironments. Second, AI-based intoxication detection models are developed using Machine Learning and Deep Learning techniques to analyze facial imagery captured under real-world surveillance conditions. Extensive experiments evaluate multiple models under varying noise and disruption scenarios to ensure robustness across both low- and high-resource computational environments. The framework also incorporates explainable AI methods to improve transparency and verify that model decisions rely on meaningful facial features. Finally, the framework integrates privacy-preserving learning mechanisms through federated learning, enabling distributed model training without transferring sensitive facial images to centralized servers. This approach protects user privacy while maintaining strong detection performance across distributed monitoring nodes. These contributions establish a secure, scalable, and privacy-aware infrastructure-level system for proactive intoxication detection, supporting intelligent transportation systems aimed at improving traffic safety.7 0Item Restricted REGULATING ALGORITHMIC DISCRIMINATION UNDER THE EU AI ACT: EVALUATING BIAS MITIGATION DUTIES FOR HIGH-RISK AND GENERAL-PURPOSE AI SYSTEMS(Saudi Digital Library, 2025) ALSOMALI, ABDULAZIZ; ZIHAO, LIAlgorithmic systems now allocate work, credit, welfare and even police attention (facial recognition systems). They are not ‘neutral’ instruments; they often reproduce and amplify structural disadvantage. This dissertation asks whether the European Union’s Artificial Intelligence Act, when coupled with the Charter of Fundamental Rights and the equality acquis, can prevent and redress such discrimination. This dissertation argues that the Act is normatively necessary but only conditionally sufficient. Its risk architecture, data‑governance duties, documentation and oversight requirements, and the upstream regime for general‑purpose models supply the right legal levers. Constitutional adequacy will materialise only if implementation embeds equality law into technical practice through three cumulative conditions: (i) standards that require context‑specific metric selection justified by proportionality and the availability of less discriminatory alternatives; (ii) supervision with genuine statistical and legal capacity across the system lifecycle; and (iii) remedial pathways that convert logs and technical files into proof under burden‑shifting rules. Thus this paper turns to a functional comparison with the United States and the United Kingdom shows how adverse‑impact doctrine, discovery, and regulator‑led guidance can be harnessed without sacrificing the coherence of the EU model. Followed by Chapter 5 which sets out a concise implementation blueprint and measurable indicators. On that basis, bias mitigation is framed not as ethics, but as a legal duty by which the Act’s success must be judged.15 0Item Restricted A Novel Emoji-Aware Computational Framework for Body-Shaming Detection in Gulf Arabic Social Media Discourse(Saudi Digital Library, 2026) Albluwi, Abeer; Rizk, DominickOnline body-shaming has become a prevalent form of appearance-based harm in Gulf Arabic social media, particularly on TikTok and Instagram. In these environments, harmful content is rarely explicit. It is carried through indirect and culturally embedded forms of expression, including sarcastic religious expressions and emojis whose meaning is often context-dependent and inverted. Despite the scale of the problem, no prior Arabic NLP work had addressed body-shaming as a standalone classification task, and no dataset existed to support it. This dissertation introduces GABSD-E (Gulf Arabic Body-Shaming Dataset, Emoji-Enriched), the first annotated corpus built specifically for this task. The dataset contains 24,988 comments from TikTok and Instagram, labeled under a three-class taxonomy: Body-Shaming (BS), General Bullying (B), and Not Bullying (NB), with Fleiss' kappa = 0.87. Exploratory analysis confirmed that 42.9% of comments contain at least one emoji, and that the BS class has the highest emoji density relative to class size, establishing that emojis are structurally embedded in how body-shaming is communicated, not incidental to it. Building on this, the dissertation proposes a novel emoji-aware representation framework that treats emojis as culturally grounded semiotic units rather than preprocessing noise. The framework consists of three components: a semantic emoji tagging layer based on a manually constructed Gulf Arabic emoji dictionary; a contextual lexicon injection layer for culturally specific expressions; and a preservation-over-deletion preprocessing pipeline that reverses the standard Arabic NLP practice of emoji removal. Five Arabic transformer models were fine-tuned and evaluated under this framework. SaudiBERT achieved the best performance, with a mean Macro-F1 of 0.9519 ± 0.0030 and accuracy of 95.27% ± 0.30 across five random seeds, substantially above all classical baselines (best classical Macro-F1: 0.6906).Ablation results confirmed that emoji removal caused the largest per-condition performance drop (−0.0589 Macro-F1), with consistent degradation observed across all classes, indicating that emoji signals provide complementary contextual cues that improve discrimination across categories. Semantic enrichment was the most critical component for the BS class (ΔF1-BS = +0.0775). An exploratory benchmark of four large language models showed that the best LLM under few-shot prompting (GPT-4o, Macro-F1 = 0.9278) still fell 0.0279 points below the fine-tuned SaudiBERT, confirming that model scale alone does not substitute for culturally grounded domain-specific modeling. The dissertation contributes GABSD-E as a reusable dataset and benchmark framework, an annotation methodology applicable to related harm categories, and a representation framework that generalizes to any Arabic NLP task where emoji pragmatics carry discriminative weight.19 0