Journal of Engineering and Emerging Technologies https://journals.cspc.edu.ph/index.php/jeet <p><span style="font-size: 0.875rem; -webkit-text-size-adjust: 100%;">The Journal of Engineering and Emerging Technologies (JEET) is a high quality, open access and international refereed journal which aims to publish original research papers in areas of engineering and technology, including Civil Engineering, Mechanical Engineering, Electrical Engineering, Electronics Engineering, Computer Science and Engineering, Information Systems and Technologies. This also covers multidisciplinary and interdisciplinary studies on the fields of engineering and technology education, management and development.</span></p> <p><span style="font-size: 0.875rem; -webkit-text-size-adjust: 100%;"><strong>ISSN</strong>: 2782-9421</span></p> en-US <p>This article is licensed under an Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) License, which permits use, distribution, and reproduction in any medium, provided that the article is properly cited, the use is non-commercial, and no modifications or adaptations are made. To view a copy of this license, visit <a href="https://creativecommons.org/licenses/by-nc-nd/4.0/">https://creativecommons.org/licenses/by-nc-nd/4.0/</a></p> [email protected] (CSPC-CRD Research Publication Management) [email protected] (Elena Clarexe Gulane) Fri, 12 May 2023 00:00:00 +0800 OJS 3.3.0.11 http://blogs.law.harvard.edu/tech/rss 60 From ChatGPT-3 to GPT-4: A Significant Advancement in AI-Driven NLP Tools https://journals.cspc.edu.ph/index.php/jeet/article/view/188 <p>Recent improvements in Natural Language Processing (NLP) have led to the creation of powerful language models like Chat Generative Pre-training Transformer (ChatGPT), Google’s BARD, Ernie which has shown to be very good at many different language tasks. But as language tasks get more complicated, having even more advanced NLP tool is essential nowadays. In this study, researchers look at how the latest versions of the GPT language model(GPT-4 &amp; 5) can help with these advancements. The research method for this paper is based on a narrative analysis of the literature, which makes use of secondary data gathered from previously published studies including articles, websites, blogs, and visual and numerical facts etc. Findings of this study revealed that GPT-4 improves the model's training data, the speed with which it can be computed, the flawless answers that it provides with, and its overall performance. This study also shows that GPT-4 does much better than GPT-3.5 at translating languages, answering questions, and figuring out how people feel about things. The study provides a solid basis for building even more advanced NLP tools and programmes like GPT-5. The study will help the AI &amp; LLM researchers, NLP developers and academicians in exploring more into this particular field of study. As this is the first kind of research comparing two NLP tools, therefore researchers suggested going for a quantitative research in the near future to validate the findings of this research.</p> Md. Saidur Rahaman, M. M. Tahmid Ahsan , Nishath Anjum, Harold Jan R. Terano, Md. Mizanur Rahman Copyright (c) 2023 Journal of Engineering and Emerging Technologies https://creativecommons.org/licenses/by-nc-nd/4.0 https://journals.cspc.edu.ph/index.php/jeet/article/view/188 Thu, 11 May 2023 00:00:00 +0800