Identify User Intention for Recommendation Using Chain-of-Thought Prompting in LLM

It is crucial for a recommendation system to accurately identify a user’s intention since it dictates the user’s selection once multiple candidates are faced. Traditional user intention identification uses the user’s selection among various items. This technique relies primarily on historical behavioural data, resulting in problems such as the cold start, unintended choice, and failure when unseen items occur. Motivated by recent advancements in Large Language Models (LLMs) like ChatGPT, we present an approach for user intention identification by leveraging Chain-of-Thought (CoT) prompting in an LLM. We use the initial user profile as inputs to LLMs and design a collection of prompts to align the LLM’s response through various recommendation tasks encompassing rating prediction, search and browse history, user clarification, etc. to identify the user’s intention and then the same input with generated user intention feed to the LLM to produce recommendations. We tested our approach with real-world datasets to demonstrate the improvements in the recommendation by comparing the recommendation without the intention of merging it into a user model. The results indicate clear improvements.