A Conceptual Framework for Conversational Search and Recommendation: Conceptualizing Agent-Human Interactions During the Conversational Search Process

TL;DR

This paper addresses how to formalize agent–human interactions in the conversational search process by proposing a conceptual framework that models the evolution of information needs through representations $CIN$, $PIN$, and $AIN$. It catalogs concrete user and agent actions (e.g., reveal, elicit, summarize) and outlines agent decision tasks (dialog policy, intent extraction) that steer the dialogue toward satisfying information needs. Building on existing dialogue-management taxonomies and theoretical frameworks for conversational search, the work provides a structured starting point to guide research, development, and evaluation of conversational search agents. The contributions offer a non-exhaustive taxonomy of actions and decisions, along with a roadmap for empirical validation and implementation, aiming to enable principled design of more effective, less cognitively taxing conversational search systems.

Abstract

The conversational search task aims to enable a user to resolve information needs via natural language dialogue with an agent. In this paper, we aim to develop a conceptual framework of the actions and intents of users and agents explaining how these actions enable the user to explore the search space and resolve their information need. We outline the different actions and intents, before discussing key decision points in the conversation where the agent needs to decide how to steer the conversational search process to a successful and/or satisfactory conclusion. Essentially, this paper provides a conceptualization of the conversational search process between an agent and user, which provides a framework and a starting point for research, development and evaluation of conversational search agents.