Trend Analysis Through Large Language Models

The rapid development of Artificial Intelligence (AI) technologies has outpaced the preparedness of many organizations. One challenge lies in the impracticality for individual researchers to sift through numerous research papers published annually. This study introduces a novel approach that combines n-gram analysis with Large Language Models (LLMs) for backtesting. The combined approach uses n-gram analysis from a corpus and then applies the LLMs to verify results. Example results are presented using a dataset of abstracts taken from the Web of Science. The implications of this study suggest a shift in how research trends are identified, enabling a more dynamic and responsive approach to research funding and investigation in the rapidly evolving field of AI.