Towards Group Decision Support with LLM-based Meeting Analysis

In many situations, groups of people need to collaborate to achieve a shared goal or solve a common problem. However, reaching an agreement can be inefficient, and critical perspectives can be overlooked during lengthy discussions. To improve this situation, this paper introduces a practical approach that uses an LLM to analyze recorded group discussions and provide informed recommendations. It analyzes meeting transcripts to identify discussed options, summarize outcomes, track decision dynamics, and generate helpful recommendations. This automation could save time, enhance transparency, and improve productivity. Through real-world case studies, we evaluate the approach to explore the strengths and limitations of using LLMs to support group decision-making.