Saudi Cultural Missions Theses & Dissertations
Permanent URI for this communityhttps://drepo.sdl.edu.sa/handle/20.500.14154/10
Browse
4 results
Search Results
Item Restricted Impact of Artificial Intelligence Integration in Emergency Department Triage on Waiting Times: A Systematic Review Compared to Conventional Practices in ED Triage.(The University of Sheffield, 2024-09) Alhazmi, Mohammed; Miles, JemieBackground: The global issue of increased patient waiting times in healthcare facilities is a pressing concern, as it can lead to significant patient harm due to delayed access to healthcare. This research proposes the integration of artificial intelligence into emergency department triage systems as a solution to mitigate this issue. Aims: To evaluate the impact of integrating Artificial Intelligence (AI) support tools on waiting times in Emergency Departments through a systematic review of existing literature. Design: A thorough systematic review of the literature was conducted by searching electronic databases and internet search engines, including ScienceDirect, Springer, and PubMed, as well as reference lists. Studies published from January 1, 2019, to May 25, 2024, were included. Articles that did not pertain to AI, interventions that were irrelevant to emergency departments (EDs) or did not provide outcomes related to reducing waiting times either directly or indirectly, or evaluation data were excluded to ensure the quality and relevance of the included studies. Results: The analysis included ten peer-reviewed journals published after January 2019 on integrated Artificial Intelligence (AI) with emergency department triage. Recent findings suggest that integrating artificial Intelligence (AI) models into the emergency department (ED) triage processes can hold significant potential for reducing overcrowding and minimising wait times. Some studies have found that AI reduces waiting times by between 20 seconds and 30 minutes. However, a study found AI to increase waiting times for categories 3 to 5 by 2.75 to 5.33 minutes. Conclusions: This review has highlighted AI's potential to bring innovative solutions to emergency department settings. Implementing these AI-driven solutions has shown promise in enhancing healthcare delivery in the emergency department. However, further research is crucial to refine these models and ensure their practical application, underscoring the importance of continued involvement in the field.44 0Item Restricted EXPLORING THE TRANSFERABILITY OF ADVERSARIAL EXAMPLES IN NATURAL LANGUAGE PROCESSING(Texas A&M University-Kingsville, 2024-06-21) Allahyani, Samah; Nijim, MaisIn recent years, there has been a growing concern about the vulnerability of machine learning models, particularly in the field of natural language processing (NLP). Many tasks in natural language processing, such as text classification, machine translation, and question answering, are at risk of adversarial attacks where maliciously crafted inputs can cause them to make incorrect predictions or classifications. Adversarial examples created on one model can also fool another model. The transferability of adversarial has also garnered significant attention as it is a crucial property for facilitating black-box attacks. In our comprehensive research, we employed an array of widely used NLP models for sentiment analysis and text classification tasks. We first generated adversarial examples for a set of source models, using five state-of-the-art attack methods. We then evaluated the transferability of these adversarial examples by testing their effectiveness on different target models, to explore the main factors such as model architecture, dataset characteristics and the perturbation techniques impacting transferability. Moreover, we extended our investigation by delving into transferability-enhancing techniques. We assisted two transferability-enhancing methods and leveraged the power of Large Language Models (LLM) to generate natural adversarial examples that show a moderate transferability across different NLP architecture. Through our research, we aim to provide insights into the transferability of adversarial examples in NLP, and shed light on the factors that contribute to their transferability. This knowledge can then be used to develop more robust, and resilient, NLP models that are less susceptible to adversarial attacks; ultimately, enhancing the security and reliability of these systems in various applications.9 0Item Restricted Automating Agency-Client Matching: Leveraging Recommender Systems for Efficient and Accurate Recommendations(University College London, 2023-12-01) Kurdi, Reem; Tanveer, UmairIn today’s fast-paced and dynamic business landscape, optimizing operational processes is paramount. Agency-client matching is an important procedure that plays a critical role in aligning agencies with client needs for successful collaborations. Traditionally, this matching process required labour-intensive, manual evaluations of numerous agencies and their portfolios. However, the advent of advanced technologies and data-driven approaches has introduced recommender systems as valuable tools to streamline and automate this process. This study presents a unique and innovative cluster-based, hybrid filtering recommender system that utilizes machine learning algorithms and data analysis for agency-client matching. The recommender system follows a comprehensive three-step process, starting with brief preparation, then topic modelling and finally, agency ranking and scoring. Firstly, the briefs undergo a comprehensive pre-processing process to ensure inclusion of relevant text data by removing irrelevant information, such as stop words and entity names. Secondly, the filtered briefs go through topic modelling using the BERTopic framework to extract the keywords and underlying themes. Briefs are first transformed into numerical vectors using BERT embeddings, which helps to capture their semantic meaning and context. After that, dimensionality reduction is applied using UMAP to cluster related briefs. As a subsequent step, DBSTREAM is applied to assign new briefs to existing clusters, or create new clusters. The final step in this block is the implementation of c-TF-IDF which helps generate topic representations by identifying the most frequent words within each topic. Lastly, based on the unique cluster identifier assigned to the new brief, agencies are ranked and scored in line with the brief’s content and client requirements. All in all, the main focus of this study is to develop a ranking and scoring algorithm, tailored with certain criteria, to effectively shortlist relevant agency options and automate the agency-client matching process.13 0Item Restricted Enhancing Clarity and Readability in Scientific Writing: An Automated Approach to Identifying Shapeless Sentences(Saudi Digital Library, 2023-11-02) Kamal, Ayah; Lopez, AdamEffective communication is essential in academic writing, where clear and coherent writing ensures research findings are disseminated effectively. However, conveying complex concepts in a readable manner remains a challenge in scientific writing. This thesis investigates automating the application of principles from the book Style: Lessons in Clarity and Grace by Williams [32] to improve the readability of scientific writing. The research focuses on identifying “shapeless” sentences that lack structure and clarity. A dataset of scientific sentences sourced from the Elsevier OA Corpus was manually annotated as “Structured”, “Shapeless” or “N/A” based on principles from Style. A Large Language Model, LLaMA-2, was fine-tuned on this dataset to classify the sentences. Optimization techniques like QLoRA enabled efficient fine-tuning within resource constraints. While, prompt engineering and few-shot learning were used to optimize inference. The fine-tuned model achieved promising accuracy in distinguishing between “Structured” and “Shapeless” sentences. The research demonstrates potential for using fine-tuned language models to automate the application of stylistic principles and enhance scientific writing. Further work is needed to expand the dataset, refine definitions, and optimize model training. Overall, this thesis establishes a foundation for using language models to identify problematic sentences and improve readability14 0