Authors: Kanan Bajaj, Savita Baerda, Manni Kumar
Abstract: This research paper aimed to find out various things ChatGPT could do. Especially in the areas of reasoning, healthcare, and education, the authors observed how ChatGPT’s learning was more personal. It adapted to student needs, offering a personalized learning experience. This truly drew students into the equation. They found that ChatGPT’s skills in logical reasoning were effective for solving problems and performing critical thinking tasks. However, a large part of their research focused on healthcare. Here, they placed ChatGPT alongside other AIs such as Gemini and Copilot. They discussed how each of them was managed in terms of diagnostic accuracy and interac-tion. There were also some differences in specialization ChatGPT performed well with general medical questions and maintained coherent conversations. Nevertheless, Gemini and Copilot generally worked better when it came to specific medical operations because those programs were designed specifically for that purpose. The paper further explored how exactly ChatGPT operated in detail. It employed modern NLP techniques and certain methods from the field of machine learning. The authors believed its evolving design made it capable of handling numerous topics. However, they also explained that it had some moderate limitations at the time, and how it could potentially be im-proved in the future. Finally, they offered a brief overview of the strengths and weaknesses of ChatGPT depending on the area of application. Hopefully, it provided some valuable insights into where it stood in the ever-expanding li-brary of AI solutions.
DOI: http://doi.org/10.5281/zenodo.16605610
