Towards Effective Modeling and Exploitation of Search and User Context in Conversational Information Retrieval
Praveen Acharya · 2023
Conversational information retrieval has garnered considerable attention in recent years. A major challenge in conversational search is formulating the most effective query during the dialogue between the searcher and the conversational agent. Unlike traditional information retrieval systems that assume users can independently create queries, conversational settings allow agents to assist users in query formulation. This alleviates the burden on users by leveraging the multi-turn nature of the conversation to aid them in reaching their information goals. Conversational context plays a vital role in the query process. In this work, we focus on understanding and leveraging conversational context from two dimensions: conversational history and knowledge history. The goal is to identify and model the relevant portions from the search dialogue and the knowledge history and to use this within the search process to improve the overall performance of conversational information retrieval systems.