Assessment of various ChatGPT models in performing SWOT analysis

The goal of this thesis is to evaluate the performance of various ChatGPT models in executing SWOT analysis task. Theoretical part is focusing on basic introduction of artificial intelligence and its branches with focus on machine learning. Later I introduce the concepts of transformer neural network which is the core architecture of most of the large language models. Last section of theoretical part is dedicated to large language models from perspective of their development throughout the time, types, and example their current usage in finance. The practical part is introducing assessment methodology for subjective tasks, where objective truth is not present. This methodology is further applied to evaluate performance of available GPT models and customized GPT model built with GPTs function on SWOT analysis of Intel company. Later I introduce the tool based in excel that significantly helps with development of customized GPT models with GPTs.