Saudi Cultural Missions Theses & Dissertations
Permanent URI for this communityhttps://drepo.sdl.edu.sa/handle/20.500.14154/10
Browse
26 results
Search Results
Item Restricted ADVANCED LARGE LANGUAGE MODEL APPROACHES AND NATURAL LANGUAGE PROCESSING TECHNIQUES TO IMPROVE HATE SPEECH DETECTION USING AI(University of Central Florida, 2025) Almohaimeed, Saad; Boloni, LadislauThe proliferation of hate speech on social networks can create a significant negative social effect, making the development of AI-based classifiers that can identify and characterize different types of hateful speech in messages highly important for stakeholders. While this is a highly challenging problem, recent advances in language models promise to advance the state of the art such that even subtle and indirect forms of hate speech can be detected. In this dissertation we present a series of contributions that improve different aspects of hate speech classification. We developed THOS, a hate speech dataset consisting of 8.3k tweets. Compared to previous datasets, THOS contains fine-grained labels that identify not only whether a tweet is offensive or hateful, but also the target of the hate. Using this dataset, we studied the degree to which finer grained classification of tweets is feasible. In the follow-up work, we focus on the difficult problem of implicit hate speech, where hate is conveyed through subtle verbal constructs and allusions, without the use of explicitly offensive terms. We evaluate the efficacy of lexicon-based methods, transfer learning, and advanced LLMs such as GPT-4 on this problem. We found that the proposed techniques can boost the detection performance of implicit hate, although even advanced models often struggle with certain interpretations. In our third contribution, we introduce the Closest Positive Cluster (CPC) auxiliary loss, which improves the generalizability of classifiers across a wide range of datasets, resulting in enhanced performance for both explicit and implicit hate speech scenarios. Finally, given the scarcity of implicit hate speech datasets and the abundance of explicit hate datasets, we proposed an approach to generalize explicit hate datasets for the classification of implicit hate speech. Additionally, the proposed approach addresses noisy label correction issues commonly found in crowd-sourced datasets. Our method comprises three key components: influential sample identification, reannotation, and augmentation. We show that the approach improves generalization across datasets and enhances implicit hate classification.15 0Item Restricted Disinformation Classification Using Transformer based Machine Learning(Howard University, 2024) alshaqi, Mohammed Al; Rawat, Danda BThe proliferation of false information via social media has become an increasingly pressing problem. Digital means of communication and social media platforms facilitate the rapid spread of disinformation, which calls for the development of advanced techniques for identifying incorrect information. This dissertation endeavors to devise effective multimodal techniques for identifying fraudulent news, considering the noteworthy influence that deceptive stories have on society. The study proposes and evaluates multiple approaches, starting with a transformer-based model that uses word embeddings for accurate text classification. This model significantly outperforms baseline methods such as hybrid CNN and RNN, achieving higher accuracy. The dissertation also introduces a novel BERT-powered multimodal approach to fake news detection, combining textual data with extracted text from images to improve accuracy. By lever aging the strengths of the BERT-base-uncased model for text processing and integrating it with image text extraction via OCR, this approach calculates a confidence score indicating the likeli hood of news being real or fake. Rigorous training and evaluation show significant improvements in performance compared to state-of-the-art methods. Furthermore, the study explores the complexities of multimodal fake news detection, integrat ing text, images, and videos into a unified framework. By employing BERT for textual analysis and CNN for visual data, the multimodal approach demonstrates superior performance over traditional models in handling multiple media formats. Comprehensive evaluations using datasets such as ISOT and MediaEval 2016 confirm the robustness and adaptability of these methods in combating the spread of fake news. This dissertation contributes valuable insights to fake news detection, highlighting the effec tiveness of transformer-based models, emotion-aware classifiers, and multimodal frameworks. The findings provide robust solutions for detecting misinformation across diverse platforms and data types, offering a path forward for future research in this critical area.34 0Item Restricted A Systematic Review of User Consent, Transparency, and Secure Data Transmission and Storage(University of Technology Sydney (UTS), 2024-11-03) Alharbi, Sultanah; Hussain, Farookh KhadeerSmart home technology is revolutionizing residential environments by connecting devices to enhance comfort, safety, and energy efficiency. However, these advancements raise significant privacy concerns, particularly in data collection, transmission, and storage. This systematic review examines user consent, transparency, and secure data handling in smart homes, identifying challenges and innovative solutions such as blockchain and AI integration. The review highlights deficiencies in current consent mechanisms, the complexity of GDPR compliance, and practical barriers to implementation, offering insights for future research and practical privacy frameworks.30 0Item Restricted Epigenetic Habitats : Mimesis and Living Architecture in Light of Catharine Malabou’s Meditation About Synaptic Chips(Univeristy College London, 2024) Alangari, Nujud; Vivaldi, Jordi4 Mimesis has been integrated with architecture for a long time—from ancient civilisations e.g. ancient Greece and the Renaissance to the modern and postmodern eras. These architectural eras tend to respond to Platonic or Kantian schemes, illustrating the evolution of architectural mimesis. For Plato, mimesis meant copying and reproducing nature through art; for Kant, however, it was more about harmonising beauty and function than copying from nature. Kant believed that art is a creation of genius which does not copy nature directly but rather reinvents nature’s rules into artistic expression. While rich in their interpretation of imitation, both concepts lack the dynamic meaning of mimesis when it comes to mimicking human intelligence. In this context, I would like to address the following question: Is the arrival of AI and robotics in architecture demanding a new epigenetic scheme for thinking about mimesis? I would like to address this question by considering Catherine Malabou’s interpretation of the concept of ‘synaptic chips’ that has been discussed in her work on epigenetic mimesis—an idea that transforms the entire picture of AI in architecture. The discussion of synaptic chips as presented by Malabou serves as a metaphorical basis for the evolution and adaptation of architectural design. Architectural designs may similarly evolve through the influence of connections that are synaptic-like; such structures respond to changes in their environments based on environmental stimuli. This approach—which is epigenetic—to mimesis suggests a shift more profound from just replicating forms to creating architectures that learn from their surroundings, thus adapting to them. This reveals a more complex interplay between form, function and environment than what is traditionally understood under Platonic or Kantian mimesis. Through this extension of mimesis by Malabou using neuroscience plus epigenetics, one can infer an avenue towards dynamic designs: designs that are more responsive and, in turn, enhance mimetic capabilities of AI systems within architecture—thereby also enhancing the architectural design’s adaptability and functionality.14 0Item Restricted THE ROLE OF ARTIFICIAL INTELLIGENCE IN ENHANCING KPIS AND OPTIMIZATION OF HUMAN RESOURCE MANAGEMENT(The Hague University, 2024-09-28) Alsamhan, Khulud; Le Fever, HansMANAGEMENT SUMMARY This thesis explores how artificial intelligence (AI) can enhance human resource management (HRM), particularly in recruitment and onboarding. The study focuses on LinkedIn's AI tools, aiming to understand their effectiveness in improving key performance indicators (KPIs) and optimizing HR processes. The research draws on a broad literature review, examining the evolution of AI in HR. AI has shown potential in automating tasks like candidate screening and onboarding, but there are challenges, including biases in AI systems and the need for continuous improvement. Using Saunders' research onion framework, a mixed-methods approach was adopted, combining surveys and interviews with HR professionals who use LinkedIn's AI tools. This approach provided a comprehensive view of AI's impact on HRM. The results indicate that AI tools significantly enhance effectiveness by automating repetitive tasks and improving candidate matching, thus reducing the time-to-hire and increasing accuracy. However, some challenges remain, such as occasional inaccuracies and the need for better user training. It's clear that refining AI algorithms and incorporating human oversight can help address these issues. In onboarding, AI tools have been successful in automating administrative tasks and personalizing the onboarding experience. Feedback suggests that AI-driven processes help new hires feel more supported and prepared. The study concludes with recommendations for further research and practical steps for implementation. It highlights the need for ongoing refinement of AI tools, better integration practices, and comprehensive training for HR professionals. Future research should focus on long-term impacts and best practices for AI in HRM. In summary, AI has the potential to transform HRM by enhancing KPIs and optimizing processes. However, a balanced approach that combines technology with human judgment is essential for maximizing these benefits. This thesis provides a foundation for future advancements in using AI in HRM.24 0Item Restricted Early Prediction of Cancer Using Supervised Machine Learning: A Study of Electronic Health Records From The Ministry of National Gurad Health Affairs(University College London (UCL), 2024-08) Alfayez, Asma; Lai, Alvina; Kunz, HolgerEarly detection and treatment of cancer can save lives; however, identifying those most at risk of developing cancer remains challenging. Electronic health records (EHR) provide a rich source of "big" data on large patient numbers. I hypothesised that in the period preceding a definitive cancer diagnosis, there exist healthcare events, such as a history of disease, captured within EHR data that characterise cancer progression and can be exploited to predict future cancer occurrence. Using longitudinal phenotype data from the EHR of the Ministry of National Guard Health Affairs, a large healthcare provider in Saudi Arabia, I aimed to discover health event patterns present in EHR data that predict cancer development in periods prior to diagnosis by developing predictive models using supervised machine learning (ML) algorithms. I used two different prediction periods: six months and one year prior to cancer diagnosis. Initially, the thesis focused on the prediction of both malignant and benign neoplasms, before moving on to predicting the future risk of malignant neoplasms (cancer), since predicting life-threatening illness remains the most important clinical challenge. To refine the approach for specific cancer types, predictive models were built for the top three malignancies in this population: breast, colon, and thyroid cancers. ML predictive models were developed using the following algorithms: (1) logistic regression; (2) penalised logistic regression; (3) decision trees; (4) random forests; (5) gradient boosting; (6) extreme gradient boosting; (7) k-nearest neighbours; and (8) support vector machine. Model performance was assessed using k-fold cross-validation and area under the curve—receiver operating characteristics (AUC-ROC). After developing different models, their performance was compared with and without hyperparameter tuning using tree-based pipeline optimization (TPOT) and GridSearch. This study provides novel proof-of-principle that ML algorithms can be applied to EHR data to develop models that can be used to predict future cancer occurrence.29 0Item Restricted TESTING COSMETIC PRODUCTS ON ANIMALS(Solent University, 2024-08) Almutiri, Hanan; Hegarty, SebastianeAbstract Technological innovation particularly the use of artificial intelligence (AI) has brought a revolution in the cosmetics and skincare industry. This paper aims to provide an understanding of how AI is being utilized in the formulation and tailoring of cosmetics and skincare products and its effects on creative product development, product safety, and consumer satisfaction. New generation technologies that include the use of artificial intelligence including machine learning, in silico modelling and virtual try-on have helped revolutionize product development and improve the delivery of customized skincare solutions and therefore customer experiences. These technologies facilitate accurate tuning of formulation and more accurate prediction of how the product will behave on different skin types, enhancing the product’s performance and hence the consumers’ confidence. The research follows the systematic review methodology and follows the PRISMA guidelines to conduct an exhaustive and transparent analysis. For the purpose of this study, a total of fifteen relevant articles were chosen in order to establish a strong basis for assessing the current and future advancements of AI in the cosmetics industry. The main issues that have been highlighted include ethical concerns, legal issues, the development of new AI methods, and the shift from animal testing to AI-based solutions. Ethical issues are mainly on the reduction of animal testing, which is viewed as both scientifically meaningless and ethically wrong. AI presents potential solutions that are compatible with the current ethical practices and emphasize in vitro and in silico experiments that are not only ethical but also effective. Regulatory issues and possibilities are discussed, with a focus on the need for new guidelines that can facilitate the implementation of AI solutions while maintaining safety and legal requirements. AI advancements in safety assessment are achieved through the establishment of models and simulations that provide improved accuracy and reliability. Based on the findings of this study, AI has the ability to revolutionize the cosmetics industry, however, more studies are required to improve on the current challenges and harness the potential of the technology. Further research should be directed towards the collection of primary data and the empirical assessment of the applicability of AI-based approaches in compliance with legal requirements.30 0Item Restricted USER MODELLING AND ADAPTIVE INTERACTION ON INTERACTIVE DASHBOARDS(University of Manchester, 2024-06-06) Alhamadi, Mohammed; Vigo, MarkelInteractive information dashboards are data visualisation tools that enable interaction with complex underlying datasets using visualisations such as charts and maps typically on a single display. The popularity of dashboards has grown across key sectors such as healthcare, education and energy, driven by the abundance of available data. Still, users face various challenges when interacting with dashboards, ranging from insufficient support for essential functionalities such as data-detail adjustment to problems with data presentation such as information overload. These problems subject users to high cognitive demands, complicate information retrieval and increase the risk of arriving at incorrect conclusions, ultimately leading to erroneous decision-making. Dashboard issues are sometimes due to developers prioritising aesthetics over functionality. At other times, they arise from a mismatch between users' visual literacy level expected by dashboard developers and the actual level of the users. When dashboard users encounter interaction problems, they exhibit certain interaction strategies as workarounds to overcome the problems. Modelling user behaviour on dashboards can shed light on these workarounds especially when applied in problematic situations. Strategies employed by users in response to interaction problems have, to a large extent, not been thoroughly explored. This thesis addresses this gap by identifying the interaction and information presentation problems faced by dashboard users, adaptation techniques that could address these problems and user strategies applied in response to problems. Results of a literature review and an interview study highlighted various problems faced by users, and at times, a disconnect between problems, adaptations and strategies. Subsequently, an experiment was conducted to identify user strategies indicative of problems when encountering four established interaction and information presentation problems: information overload, inappropriate data order \& grouping, ineffective data presentation and misaligned visual literacy expectations. These problems were prioritised based on their severity and the limited understanding of user strategies when encountering them. We found clear distinctions between the strategies applied on problematic and adapted dashboards. Then, we incorporated the strategies, along with graph literacy, in user models to predict usability. In a final user study, we ecologically validated the effect of the majority of the influential user strategies on usability in real-world dashboards. While filtering data was linked to negative outcomes, customisation made users more effective. Encouragingly, usability predictions were more accurate on problematic dashboards and challenging tasks. These promising results open up avenues for tailored interventions to address the problems in real time.23 0Item Restricted Impact of Strategic Knowledge Management Practices on ERP Systems in Saudi Arabia Business Organizations(Hunan University, 2024-06-15) Baslom, Mohammed Majdy M; Shu, TongMany organizations are currently implementing enterprise resource planning (ERP) to address their operational challenges. Despite its appeal, ERP implementation is fraught with obstacles and complications, particularly in developing nations. Recent studies indicate that the implementation of EM-ERP has significantly enhanced production, services, revenues, and employee well-being. Both developed and developing nations have witnessed the emergence of novel management levels and innovative concepts. The "Saudi Vision 2030" initiative is a significant national undertaking with substantial economic implications for Saudi Arabia. Knowledge management (KM) is assuming new, crucial responsibilities in advancing the industrial business environment, especially in the face of globalization and intense corporate competition. Organizations are increasingly focusing on the development and application of knowledge as a strategic asset. In 2017, the industrial sector contributed approximately 45% of Saudi Arabia's gross domestic product (GDP), a figure expected to rise as KM-ERP programs are integrated into Saudi business organizations, particularly in the manufacturing sector. This Dissertation investigates the critical factors influencing the adoption of ERP systems for effective KM in the Saudi Arabian manufacturing sector. The study aims to determine how KM can be utilized as a strategic resource to optimize ERP systems, consequently enhancing organizational competitiveness. Additionally, it assesses the role of support teams, providing a novel perspective on how human resources and team interactions can substantially influence ERP and KM processes. The integration of ERP systems and KM is essential for improving the performance, efficiency, and competitiveness of industrial businesses in Saudi Arabia. ERP systems automate and integrate business operations, including human resources, accounting, inventory, production, and sales. KM connects the generation, dissemination, and implementation of knowledge within an organization to achieve its goals. Thus, it is crucial for manufacturing companies to develop and implement contemporary strategies. Given the global impact of AI on research and implementation, expanding Saudi Arabia's research program is vital. By monitoring and analyzing data from machinery and shop floor processes, manufacturers can detect patterns to predict or prevent malfunctions. ERP systems are critical digital infrastructures that link operations throughout manufacturing enterprises. With the rapid development of AI capabilities, ERP platforms are poised for transformation. The integration of intelligent features can provide unprecedented connectivity, visibility, efficiency, and insight, revolutionizing the manufacturing sector in Saudi Arabia and enhancing the nation's economic status. This investigation achieves several essential contributions. First, it identifies critical factors influencing the success or failure of an ERP-KM environment within Saudi Arabian manufacturing organizations. The study focuses on organizational learning readiness, change management, ERP adoption scenarios, and KM methods used by Saudi enterprises. Second, it integrates information management and decision-making by examining knowledge alignment, collaboration, and communication. The study uses quantitative methods, including logistic regression and partial least squares SEM, followed by CFA and structural model assessment using Python. Third, it evaluates the impact of ERP and KM systems on support teams within business organizations, quantifying this impact with statistical metrics such as goodness of fit, R-squared, Chi-square, RMSEA, CFI, and TLI. The study identifies adoption barriers and explores how social, political, economic, and cultural factors influence KM and ERP implementation. Lastly, the research implements the PLS-SEM model and demonstrates that strategic business information distribution significantly impacts AI awareness in KM. It highlights the necessity of instruction and training in novel technologies and examines the role of learning environments and AI awareness in organizational structures. By exploring interdepartmental collaboration and information exchange, it provides a comprehensive perspective on organizational dynamics impacting ERP and KM systems. Incorporating strategic KM practices into the ERP systems of Saudi Arabian manufacturing companies will optimize ERP capabilities and reveal both financial and non-financial benefits. This Dissertation contributes to organizational learning readiness, change management, and technological adoption, providing insights into the optimization of ERP and KM systems in Saudi Arabia.28 0Item Restricted Digital Technologies in Accounting Firms: Adoption, Impact and New Avenues for Future Research(University of East Anglia, 2023-06) Alsahlawi, Saja; Guven-Uslu, Pinar; Dewing, IanPaper 1 Purpose – The purpose of this paper is to review the literature on the impact of digital technologies on accounting practice and accountants' roles in the context of accounting firms. Design/methodology/approach – A scoping review of academic studies was used to achieve the study's purpose. Findings – Only thirteen empirical papers on the impact of digital technologies on accounting practice and accountants' roles in the context of accounting firms were identified. Furthermore, the review revealed that discussion papers and anecdotal claims dominate the literature. It is important for future research to consider to what extent the accounting profession, and accounting firms in particular, are embracing digital technology and how it is impacting accounting practice and accountants' roles. The findings also reveal that the challenges and risks associated with digital technologies are unaccounted for and ignored in the literature. Originality/value – This paper contributes to the field of accounting research by providing an overview of emergent literature on the usage of digital technologies and its impact on accounting and accountants' roles in accounting firms context. It is also the first to synthesise and discuss the challenges/risks, as well as the opportunities/benefits associated with digital technologies within that context while also aiming to serve as a catalyst for future research.140 0
- «
- 1 (current)
- 2
- 3
- »