SACM - United Kingdom
Permanent URI for this collectionhttps://drepo.sdl.edu.sa/handle/20.500.14154/9667
Browse
4 results
Search Results
Item Restricted Exploring the Efficacy of ChatGPT Feedback in L2 Writing: Perspectives from Teachers and Students(Lancaster University, 2024) Alamri, Tahani; Lo, JustinThe emergence of user-friendly, widely accessible generative AI tools like ChatGPT has sparked widespread interest in academia. As students increasingly turn to these tools for various tasks, educational institutions have adopted different policies, from complete bans to active exploration of their potential applications. One frequently proposed application in recent research is using these tools to provide written corrective feedback on students' writing. This dissertation investigates the efficacy of ChatGPT feedback and the potential of AI tools in L2 writing instruction through the perspectives of L2 teachers and students. The aim is to contribute to the current understanding of AI and how it could potentially benefit or hinder L2 writing instruction, as well as determine the optimal approach for incorporating this tool. To achieve this, two online questionnaires were utilized, one targeting teachers and one for students. Participants evaluated samples of GPT feedback on a 5-criteria scale and shared their insights about the potential role of GPT in writing instruction in addition to the potential benefits and concerns. The findings showed that teachers and students generally viewed GPT feedback favorably. However, they acknowledged the limitations of AI feedback and the potential challenges AI systems might create if misused. Therefore, the majority opted for a balanced approach between teacher and AI feedback for maximized benefits and reducing risks.28 0Item Restricted Saudi Arabian Students’ Perceptions of Chat GPT in Relation to their English Language Writing Skills while Studying in the UK(Newcastle University., 2024-08-30) Alsubaie, Jawaher Obaid; Seedhouse, PaulThis research investigates Saudi Arabian students' perceptions of ChatGPT and its impact on their English writing skills while studying in the UK. The study aims to explore how ChatGPT influences Saudi students’ writing development, uncover the perceived advantages and disadvantages of the tool, and provide recommendations for its integration into Saudi education in alignment with AI policies. A mixed-method research design was employed, incorporating both quantitative and qualitative approaches. Data were collected through a questionnaire administered to 270 Saudi EFL students in the UK, documents analysis, and semi-structured interviews. The findings reveal that Saudi students generally perceive ChatGPT as a beneficial tool for improving their English writing skills, particularly in terms of technical writing, idea generation, and increased confidence. However, concerns were raised about over-reliance on the tool and inaccuracies in its output. The qualitative interviews confirmed these results, shedding light on how students view ChatGPT as a supportive resource for enhancing language proficiency and efficiency, while also noting the potential for dependency and misinformation. Moreover, the analysis shows both Newcastle university and Saudi AI policies prioritize responsible AI and academic integrity but differ in focus. Saudi policies emphasise risk management, while Newcastle prioritizes academic integrity and learning. Sentiment analysis revealed Saudi's cautious approach, while Newcastle balances risks with the benefits of AI. Based on these findings, the study recommends integrating ChatGPT into Saudi educational settings with clear guidelines, proper training, and an emphasis on responsible usage. Such measures will help ensure that students can maximize the benefits of AI tools while mitigating the risks, particularly as they advance in their academic and professional careers.38 0Item Restricted Creating Synthetic Data for Stance Detection Tasks using Large Language Models(Cardiff University, 2023-09-11) Alsemairi, Alhanouf; Manchego, Fernando AlvaStance detection is a natural language processing (NLP) task that analyses people’s stances (e.g. in favour, against or neutral) towards a specific topic. It is usually tackled using supervised classification approaches. However, collecting datasets with suitable human annotations is a resource-expensive process. The impressive capability of large language models (LLMs) in generating human-like text has revolutionized various NLP tasks. Therefore, in this dissertation, we investigate the capabilities of LLMs, specifically ChatGPT and Falcon, as a potential solution to create synthetic data that may address the data scarcity problem in stance detection tasks, and observe its impact on the performance of stance detection models. The study was conducted across various topics (e.g. Feminism, Covid-19) and two languages (English and Arabic). Different prompting approaches were employed to guide these LLMs in generating artificial data that is similar to real-world data. The results demonstrate a range of capabilities and limitations of LLMs for this use case. ChatGPT’s ethical guidelines affect its performance in simulating real-world tweets. Conversely, the open-source Falcon model’s performance in resembling the original data was better than ChatGPT’s; however, it could not create good Arabic tweets compared to ChatGPT. The study concludes that the current abilities of ChatGPT and Falcon are insufficient to generate diverse synthetic tweets. Thus, additional improvements are required to bridge the gap between synthesized and real-world data to enhance the performance of stance detection models.28 0Item Restricted Medical Screening Assistant: A Chatbot to Help Nurses(Saudi Digital Library, 2023-11-08) Al Rabeyah, Abdullah Saleh; Da Silva, Rogerio E; Goes, FabricioOver the last several years, Machine Learning has emerged as a key player in the healthcare industry. The use of chatbots is a notable application of artificial intelligence within the field of healthcare. The advent of the ChatGPT revolution represents a significant breakthrough in the realm of natural language processing, a fundamental aspect of chatbot programming. This development has simplified the implementation of GPT to engage in user communication and fulfill the objectives of the application. The objective of this project is to reduce the excessive workloads faced by healthcare professionals and enhance the efficiency of decision-making processes. This will be achieved via the development of an intelligent medical chatbot as a mobile application, specifically designed to support nurses in conducting early patient diagnoses by analyzing symptoms. The chatbot uses Swift programming language for the iOS front-end and Python with Flask for the backend. It incorporates the ChatGPT API and machine learning models to effectively comprehend and interpret user inquiries. This project uses a Kaggle dataset of 41 distinct diseases along with their corresponding symptoms. The model is trained using Logistic Regression to predict the prognosis. The responsibility of managing the dialogue between the user and the chatbot, leading up to the compilation of the definitive list of symptoms shown by the patient, lies with ChatGPT. The use of a Flask RESTful API facilitates direct interaction between the iOS application and the server-side infrastructure. Finally, the application will provide the nurse with the five most probable prognoses, along with the prediction confidence scores, depending on the symptoms supplied. Additionally, the application will offer a description of the disease and provide precautionary measures for the patient.18 0