Saudi Cultural Missions Theses & Dissertations

Permanent URI for this communityhttps://drepo.sdl.edu.sa/handle/20.500.14154/10

Browse

Search Results

Now showing 1 - 10 of 19
  • ItemRestricted
    TOWARDS ROBUST AND ACCURATE TEXT-TO-CODE GENERATION
    (University of Central Florida, 2024) almohaimeed, saleh; Wang, Liqiang
    Databases play a vital role in today’s digital landscape, enabling effective data storage, manage- ment, and retrieval for businesses and other organizations. However, interacting with databases often requires knowledge of query (e.g., SQL) and analysis, which can be a barrier for many users. In natural language processing, the text-to-code task, which converts natural language text into query and analysis code, bridges this gap by allowing users to access and manipulate data using everyday language. This dissertation investigates different challenges in text-to-code (including text-to-SQL as a subtask), with a focus on four primary contributions to the field. As a solution to the lack of statistical analysis in current text-to-code tasks, we introduce SIGMA, a text-to- Code dataset with statistical analysis, featuring 6000 questions with Python code labels. Baseline models show promising results, indicating that our new task can support both statistical analysis and SQL queries simultaneously. Second, we present Ar-Spider, the first Arabic cross-domain text-to-SQL dataset that addresses multilingual limitations. We have conducted experiments with LGESQL and S2SQL models, enhanced by our Context Similarity Relationship (CSR) approach, which demonstrates competitive performance, reducing the performance gap between the Arabic and English text-to-SQL datasets. Third, we address context-dependent text-to-SQL task, often overlooked by current models. The SParC dataset was explored by utilizing different question rep- resentations and in-context learning prompt engineering techniques. Then, we propose GAT-SQL, an advanced prompt engineering approach that improves both zero-shot and in-context learning experiments. GAT-SQL sets new benchmarks in both SParC and CoSQL datasets. Finally, we introduce Ar-SParC, a context-dependent Arabic text-to-SQL dataset that enables users to interact with the model through a series of interrelated questions. In total, 40 experiments were conducted to investigate this dataset using various prompt engineering techniques, and a novel technique called GAT Corrector was developed, which significantly improved the performance of all base- line models.
    11 0
  • ItemRestricted
    Saudi Students’ Experiences Using Artificial Intelligence to Support Well-Being While Studying Abroad
    (University of Sheffield, 2024-09) Barhyem, Smer; Rowsell, Jennifer
    This desk-based study explores Saudi Students’ Experiences Using Artificial Intelligence to Support Well-Being while studying abroad. It focuses on their challenges to investigate how AI can address them, through a qualitative approach. Data were collected through secondary sources highlighting Saudi master's students' challenges studying abroad and the impact of AI on their well-being. The data were analyses by using thematic analysis to determine meaningful themes. The research discovers that AI can offer helpful solutions, by providing language support, promoting social integration, and offering mental health services. Despite the possible benefits, there are some concerns about ethical issues related to AI, such as privacy, breaches, and biases. This research seeks to encourage Saudi master's students to use AI by explaining how it offers multiple services and the impact on their well-being and suggesting some recommendations. These recommendations include the opportunity to Invest in AI support systems in the universities to enhance their language and encourage them to communicate with others and develop mental health support services, leading to starting the treatment quickly and offering academic support services. Additionally, by considering these recommendations universities can create supportive environments for Saudi master's students studying abroad, leading to improved well-being and academic success.
    26 0
  • ItemRestricted
    Exploring the Applications of Artificial Intelligence in Enhancing Pre-Hospital Care: A Scoping Review
    (Queen’s University, Belfast, 2024) Alfaifi, Yahya; Clarke, Susan
    Artificial Intelligence (AI) has the potential to significantly improve pre-hospital care, especially in emergency medical services (EMS). However, its current application remains scattered, with varying integration levels across care stages. This scoping review aims to map and assess existing research on AI applications within pre-hospital care without focusing on specific AI technologies, such as machine learning (ML), deep learning (DL), or decision support systems (DSS). The review reflects the current research landscape, capturing how AI is utilised across critical stages such as call-taking, dispatch, and on-scene assessment. Using the framework developed by Arksey and O’Malley (2005), a systematic search was conducted across multiple databases to identify studies relevant to AI in pre-hospital care. The scope was deliberately broad to capture a comprehensive view of the available literature, focusing on identifying areas where further research is needed. The findings indicate that DSS is commonly used to support decision-making in call-taking and dispatch, while more advanced AI applications like ML and DL show potential in predictive analytics and real-time decision-making. However, these technologies are still in their early stages of real-world implementation. This review highlights the gaps in AI research, particularly in the later stages of prehospital care, such as transport and handover. Further exploration is necessary to unlock AI’s full potential in enhancing EMS operations and outcomes.
    38 0
  • ItemRestricted
    AI GENERATED TEXT VS. HUMAN GENERATED TEXT
    (University of East Anglia, 2024-09) Hadi, Nedaa; Misri, Kazhan
    The ability to distinguish between AI-generated and human-generated texts is becom- ing increasingly critical as AI technologies advance. This dissertation explores the development and evaluation of various machine learning models to accurately classify text as either AI-generated or human-generated. The research aims to identify the most effective classification techniques and preprocessing methods to enhance model performance and generalization across different text datasets. A range of machine learning and deep learning models, including Support Vec- tor Machine (SVM), Random Forest, Logistic Regression, Decision Tree, BERT, and LSTM, were employed to evaluate their effectiveness in distinguishing between the two types of texts. The study utilized a balanced and representative dataset through data sampling and augmentation techniques. Key preprocessing steps were implemented to refine the input data, and hyperparameter tuning was conducted to optimize model performance. The generalization capabilities of the models were further tested on additional datasets with varying text characteristics. The findings revealed that SVM and Random Forest models achieved the highest accuracy and reliability in classifying texts, demonstrating strong performance across multiple evaluation metrics. In contrast, deep learning models like BERT and LSTM were less effective under the given conditions, suggesting a need for more extensive datasets and computational resources to leverage their full potential. These results highlight the strengths and limitations of different approaches to text classification, providing a foundation for future research to enhance AI detection in diverse applications.
    21 0
  • ItemRestricted
    The Role of Artificial Intelligence in Breast Cancer Screening Programmes: A Literature Review and Focus Upon Policy Implications
    (The University of Edinburgh, 2024) Alrabiah, Alanoud; Hellowell, Mark
    Background: Breast cancer (BC) is a leading cause of morbidity and mortality amongst older women, leading to the introduction of screening programmes to support earlier detection and improved survivability. Current screening programmes rely upon the performance of radiologists in terms of accuracy; however, evidence shows that both under and overdiagnosis means screening also results in harms to some women. Artificial intelligence is then a promising technology for improving the accuracy of mammogram screening. Aim: To describe the potential roles of AI in BC screening, and the potential benefits, limitations and risks in these roles. Methods: PubMed, SCOPUS, and CINAHL were searched. Primary research studies published in English and in the last ten years, investigating the accuracy of AI systems for screening BC, were eligible for review. Evidence was appraised using the CASP (2024) checklists and data analysed narratively. Results: 14 studies were found eligible for review, mostly adopting a retrospective study design or laboratory study design. Roles for AI in BC screening include as a standalone system replacing radiologists entirely, as risk stratification systems used before radiologist readings, or as reader aids. While some studies reported AI systems to be superior, others reported accuracy to be inferior to radiologist readings. Differences in results could be due to variations in AI system or radiologist performance. Conclusion: There is insufficient evidence to support the use of AI in BC screening programmes, and more robust, prospective studies comparing readings from clinical practice are urgently required. Policy must also be implemented to regulate the use of AI until there is sufficient evidence to support its use.
    12 0
  • Thumbnail Image
    ItemRestricted
    Knowledge, Use, and Confidence in Artificial Intelligence Applications Among Orthodontists in the UK and Ireland
    (The University of Edinburgh, 2024) Sabbagh, Abdulrahman; McGuinness, Niall
    Background: Artificial intelligence (AI) has been applied in orthodontics using different applications, including cephalometric tracing, remote and initial assessment, remote monitoring of treatment progress, and extraction decision-making. This study aims to assess knowledge, usage, confidence, and future interest in AI applications amongst orthodontists in the United Kingdom and Ireland. Materials and Methods: A cross-sectional study was conducted amongst orthodontists in the United Kingdom and Ireland. A self-reported questionnaire was used. Data was collected on participant demographics, as well as knowledge, usage, confidence, and future use of different AI applications. Pearson chi-square tests were used to assess if demographics, region of work, sector of work, and years of experience influenced responses. Results: A total of 331 responses were received. There was a general awareness that AI can be used in orthodontics in 80.4% of respondents. In addition, the overall mean knowledge, usage, and confidence levels of the examined AI applications were 51.3%, 16.6% and 18.7% respectively. Knowledge, usage, and confidence levels for specific AI applications differed, with the greatest familiarity, usage, and confidence observed in AI applications for cephalometric tracing and remote monitoring. Alternatively, the lowest awareness, usage, and confidence were attributed to AI applications that assisted in identifying the need for extractions. Additionally, most orthodontists (81%) consider AI to be beneficial for future use and the majority (96.7%) were open to learning about it. Statistically Significant associations (P >0.05) were discovered between knowledge, usage, and confidence in various AI applications and between multiple factors including healthcare sectors, practice regions, and gender. Conclusion: This study revealed differing levels of knowledge, usage, and confidence in various AI applications among practitioners in the UK and Ireland. The findings suggest a knowledge-implementation gap that might be beneficial to be targeted by educational means to increase the adoption of AI technology in the orthodontic practice.
    73 0
  • Thumbnail Image
    ItemRestricted
    The use of Artificial Intelligence in Emergency Triage versus Conventional Triage in Adult Patients Affects Emergency Department Overcrowding.
    (Univrsity of sydney, 2024-07-17) Alghamdi, Norah; Randall, Sue
    Aim: This review aims to identify whether using artificial intelligence (AI) in emergency triage versus conventional triage in adult patients affects emergency department (ED) overcrowding. Background: In emergency medicine, triage is an essential procedure for establishing patient care priorities according to severity (Defilippo et al., 2023). Typically, this practice depends on the experience and decision-making skills of emergency nurses, which can sometimes result in differences in how patients are evaluated and interruptions in receiving timely care. Due to several factors, such as ED overcrowding, the triage process can be delayed or misjudged. Overcrowding occurs when the number of patients waiting to be seen, examined, or discharged from the ED exceeds the ED's structural or personnel capacity (Cameron et al., 2009). Since the issue of human resources is a complex issue, finding alternative methods to aid the workforce such as Artificial Intelligence (AI) should be investigated. This paper aims to discover how using AI in emergency triage can reduce and accelerate the triage process. Methods: an integrative literature review was conducted via systematic research in three electronic databases, CINHAL, MEDLINE, and PubMed. The literature included studies that used artificial intelligence in triage and how AI affects emergency overcrowding. Some of these studies did not measure overcrowding directly but studied the effects of overcrowding, and how can reflect the care provided, and the time to initiate treatment. Results: AI-based triage systems, which use machine learning algorithms to forecast patient acuity, hospitalisation, and death, provide notable improvements over conventional methods. These systems reduce errors related to human judgment and cognitive biases by improving accuracy and efficiency in triage choices using clinical data and electronic health records (Lee et al., 2024; Shahbandegan et al., 2022). Conventional triage techniques, on the other hand, mostly rely on the interpretation of individual clinicians, leaving room for errors. AI-based solutions, such as those that employ extreme gradient boosting algorithms, offer real-time decision support, enhancing patient outcomes by recommending crucial treatments for prompt and suitable care in emergency rooms. When compared to conventional techniques, AI-based triage systems demonstrate higher prediction ability and the potential to improve emergency care. Conclusion: Artificial intelligence showed positive results in emergency health settings and reduced overcrowding. In addition, future research requires algorithm refinement to increase generalisability and mitigate false positive cases.
    35 0
  • Thumbnail Image
    ItemRestricted
    Machine Learning for Improved Detection and Segmentation of Building Boundary
    (Cardiff University, 2022-09-27) Algarni, Salem; Mourshed, Monjur
    This thesis addresses the need for rapid assessment of damaged assets, such as buildings, following natural or man-made disasters. Traditional manual visual analysis is labor-intensive and time-consuming, prompting the exploration of automated detection methods using multi-source geospatial data. The research reviews existing object detection methods and introduces two novel post-processing techniques. Artificial intelligence, particularly convolutional neural networks (CNNs), is employed to improve building detection accuracy. The study emphasizes the superior performance of CNN architectures, especially Region-based Convolutional Neural Network (Mask R-CNN), which enhances semantic detection and boundary recognition. Mask R-CNN, combined with conditional random fields (CRFs), effectively identifies and refines building contours in satellite images. The proposed post-processing techniques modify the relative orientation properties of buildings and integrate key points from two neural networks to adjust predicted contours using innovative snap algorithms. The results demonstrate notable improvements in boundary detection accuracy, with enhancements of 2.5% in F1-Score, 4.6% in Intersection over Union, and 1% in overall pixel accuracy compared to current state-of-the-art methods. The thesis highlights the potential of CNNs in automating image processing tasks, learning complex concepts from raw data, and aiding infrastructure planning and disaster response.
    39 0
  • Thumbnail Image
    ItemRestricted
    Artificial Intelligence Performance on Digital Mammography, Synthetic Mammography, and Digital Breast Tomosynthesis: A Systematic Review.
    (Saudi Digital Library, 2023-11-23) Alamri, Abdulrahman; Polycarpou, Irene; Feltbower, Richard
    Objective This review aims to evaluate the effectiveness of AI in breast cancer screening by examining its performance across various modalities, including digital mammography (DM), digital breast tomosynthesis (DBT), and synthetic mammography (SM), as well as its ability to detect different breast densities and types of lesions. Method The Medline and Scopus databases were searched electronically from the earliest date to 9 August 2023 for studies that met the inclusion criteria. A QUADAS-2 tool was used to assess the quality of the included studies. The diagnostic performance metrics of included studies were reported, such as sensitivity, specificity, accuracy, and the area under the curve (AUC). A narrative synthesis with tables and figures was used to present the result. Results Based on the search, 13 studies were included. Most studies were published in 2021 (46%) and in the USA (4/13). A majority of the included studies used cross-sectional datasets to test their AI tools, and the USA was found to be the most common dataset location. Two studies developed AI tools to classify breast density, and the mean accuracy was 82% for DM and 80% for SM, whereas the AUC were 0.94 and 0.93, respectively. The remaining 11 studies focused on AI diagnostic applications. Four DM studies showed a sensitivity (mean: 73.7, median: 69.9, IQR: 14), two SM studies (mean: 79.9), and five DBT sensitivity studies (mean: 82.2, median: 84, IQR: 21). Conclusion AI in breast classification has a comparable performance, while the DBT has the highest performance among modalities and techniques.
    66 0
  • Thumbnail Image
    ItemRestricted
    Barriers and Facilitators of AI-Driven Tools in Healthcare Project Management in Saudi Arabia: An Exploratory Study.
    (Saudi Digital Library, 2023-12-11) Alyami, Khawlah; Pagano, Rosane
    The aim of this research is to identify and analyse the primary barriers and facilitators that could influence the project management of healthcare AI initiatives in Saudi Arabia. By doing so, the study seeks to contribute to the existing body of knowledge and assist in formulating strategies and policies that can expedite the digital transformation process in the healthcare sector. The research employs a qualitative methodology, utilizing semi-structured interviews and focus groups as tools for data collection. The participants included healthcare project coordinators and senior project leads with substantial experience in the field. The research aims to fill a gap in the existing literature by offering insights into the unique factors impacting AI technology adoption in the Saudi healthcare sector. Findings from the study identify strategic initiatives, skill development, and collaborations as central facilitators in the integration process. Simultaneously, data management challenges, ethical and regulatory concerns, stakeholder management, and risk management are recognized as significant barriers. The study offers specific recommendations to address these barriers, focusing on improving data security and compliance with regulations. This research serves as a valuable resource for stakeholders in the healthcare sector, providing actionable insights to advance the integration of AI tools in Saudi Arabia, aligning with the objectives of Vision 2030.
    42 0

Copyright owned by the Saudi Digital Library (SDL) © 2025