Can We Predict the Next Question? A Collaborative Filtering Approach to Modeling User Behavior

TL;DR

This paper introduces a novel collaborative filtering approach that models dynamic user-question interactions to improve question prediction in dialogue systems, capturing evolving interests and behaviors.

Contribution

It proposes the CFQP framework combining personalized memory and graph-based preference propagation to better predict user questions over time.

Findings

Effective in modeling evolving user interests

Improves question prediction accuracy

Enhances adaptive dialogue system capabilities

Abstract

In recent years, large language models (LLMs) have excelled in language understanding and generation, powering advanced dialogue and recommendation systems. However, a significant limitation persists: these systems often model user preferences statically, failing to capture the dynamic and sequential nature of interactive behaviors. The sequence of a user's historical questions provides a rich, implicit signal of evolving interests and cognitive patterns, yet leveraging this temporal data for predictive tasks remains challenging due to the inherent disconnect between language modeling and behavioral sequence modeling. To bridge this gap, we propose a Collaborative Filtering-enhanced Question Prediction (CFQP) framework. CFQP dynamically models evolving user-question interactions by integrating personalized memory modules with graph-based preference propagation. This dual mechanism allows the system to adaptively learn from user-specific histories while refining predictions through collaborative signals from similar users. Experimental results demonstrate that our approach effectively generates agents that mimic real-user questioning patterns, highlighting its potential for building proactive and adaptive dialogue systems.