Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Should We Tailor the Talk? Understanding the Impact of Conversational Styles on Preference Elicitation in Conversational Recommender Systems

Ivica Kostric, Krisztian Balog, Ujwal Gadiraju

TL;DR

This paper investigates how conversational styles in CRSs influence preference elicitation, task performance, and user satisfaction. Using a controlled study in scholarly literature recommendations, it contrasts HighInvolvement and HighConsiderateness styles and introduces an Agency condition allowing style switching, across domains of varying familiarity. The authors implement the Conversational Scholarly Assistant (CSA) with modular NLU/DM/NLG components, BM25-based ranking, and LLM-assisted relevance scoring to quantify outcomes. Key findings show that style effectiveness depends on user domain familiarity, with high involvement aiding experienced users and high considerateness aiding novices; permitting style switching yields the strongest task performance. The work advances CRS design by highlighting the value of adaptive dialogue strategies and transparent user agency, and provides open-source resources for replication and extension.

Abstract

Conversational recommender systems (CRSs) provide users with an interactive means to express preferences and receive real-time personalized recommendations. The success of these systems is heavily influenced by the preference elicitation process. While existing research mainly focuses on what questions to ask during preference elicitation, there is a notable gap in understanding what role broader interaction patterns including tone, pacing, and level of proactiveness play in supporting users in completing a given task. This study investigates the impact of different conversational styles on preference elicitation, task performance, and user satisfaction with CRSs. We conducted a controlled experiment in the context of scientific literature recommendation, contrasting two distinct conversational styles, high involvement (fast paced, direct, and proactive with frequent prompts) and high considerateness (polite and accommodating, prioritizing clarity and user comfort) alongside a flexible experimental condition where users could switch between the two. Our results indicate that adapting conversational strategies based on user expertise and allowing flexibility between styles can enhance both user satisfaction and the effectiveness of recommendations in CRSs. Overall, our findings hold important implications for the design of future CRSs.

Should We Tailor the Talk? Understanding the Impact of Conversational Styles on Preference Elicitation in Conversational Recommender Systems

TL;DR

This paper investigates how conversational styles in CRSs influence preference elicitation, task performance, and user satisfaction. Using a controlled study in scholarly literature recommendations, it contrasts HighInvolvement and HighConsiderateness styles and introduces an Agency condition allowing style switching, across domains of varying familiarity. The authors implement the Conversational Scholarly Assistant (CSA) with modular NLU/DM/NLG components, BM25-based ranking, and LLM-assisted relevance scoring to quantify outcomes. Key findings show that style effectiveness depends on user domain familiarity, with high involvement aiding experienced users and high considerateness aiding novices; permitting style switching yields the strongest task performance. The work advances CRS design by highlighting the value of adaptive dialogue strategies and transparent user agency, and provides open-source resources for replication and extension.

Abstract

Conversational recommender systems (CRSs) provide users with an interactive means to express preferences and receive real-time personalized recommendations. The success of these systems is heavily influenced by the preference elicitation process. While existing research mainly focuses on what questions to ask during preference elicitation, there is a notable gap in understanding what role broader interaction patterns including tone, pacing, and level of proactiveness play in supporting users in completing a given task. This study investigates the impact of different conversational styles on preference elicitation, task performance, and user satisfaction with CRSs. We conducted a controlled experiment in the context of scientific literature recommendation, contrasting two distinct conversational styles, high involvement (fast paced, direct, and proactive with frequent prompts) and high considerateness (polite and accommodating, prioritizing clarity and user comfort) alongside a flexible experimental condition where users could switch between the two. Our results indicate that adapting conversational strategies based on user expertise and allowing flexibility between styles can enhance both user satisfaction and the effectiveness of recommendations in CRSs. Overall, our findings hold important implications for the design of future CRSs.

Paper Structure

This paper contains 19 sections, 4 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Conversational Recommender Systems
  4. Preference Elicitation
  5. Conversational Styles
  6. Study Design
  7. Conversational Styles: High Involvement and High Considerateness
  8. Pacing and Tone
  9. Content Introduction and Questioning
  10. Task Design
  11. Evaluation
  12. Conversational Scholarly Assistant System
  13. Conversational Flow
  14. System Architecture and Modeling
  15. Results and Analysis
  16. ...and 4 more sections

Figures (4)

  • Figure 1: Contrasting example conversations between two conversational styles for the same user need. HighInv style on the left and HighCon style on the right.
  • Figure 2: The user journey through the study phases: pre-task familiarization, main task interaction with the conversational system (counterbalanced with respect to which style is first encountered by participants), and post-task feedback. Participants completed three tasks in a randomized order with varying conversational styles, followed by feedback after each task.
  • Figure 3: Dialogue flow. The green path shows the main flow for the HighInv, while the orange shows the path for the HighCon conversational style.
  • Figure 4: Architecture of the Conversational Scholarly Assistant (CSA) system used in this study.