Personalized Recommendations via Active Utility-based Pairwise Sampling

2025-08-12

AI Summary

Traditional recommender systems rely on explicit ratings or complete item rankings; however, ratings are susceptible to behavioral biases and subjective noise, while obtaining full rankings is prohibitively costly and often infeasible. To address these limitations, we propose a utility-driven active preference learning framework that models user preferences via pairwise comparisons grounded in the Plackett–Luce probabilistic ranking model. Our approach jointly estimates latent utilities using hybrid representations—integrating matrix factorization with neural networks. Crucially, we directly couple query selection (i.e., which item pairs to solicit) with downstream recommendation utility, enabling task-specific utility functions. We evaluate our method on two heterogeneous real-world tasks: movie recommendation and university admissions candidate screening. Results demonstrate substantial improvements in both recommendation accuracy and annotation efficiency—achieving higher precision with significantly fewer pairwise queries—thereby validating the framework’s generality, effectiveness, and practical applicability.

Technology Category

• Machine Learning: Learning Preferences or Rankings
• Knowledge Representation and Reasoning: Preferences
• Data Mining & Knowledge Management: Recommender Systems

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs

Abstract

Recommender systems play a critical role in enhancing user experience by providing personalized suggestions based on user preferences. Traditional approaches often rely on explicit numerical ratings or assume access to fully ranked lists of items. However, ratings frequently fail to capture true preferences due to users' behavioral biases and subjective interpretations of rating scales, while eliciting full rankings is demanding and impractical. To overcome these limitations, we propose a generalized utility-based framework that learns preferences from simple and intuitive pairwise comparisons. Our approach is model-agnostic and designed to optimize for arbitrary, task-specific utility functions, allowing the system's objective to be explicitly aligned with the definition of a high-quality outcome in any given application. A central contribution of our work is a novel utility-based active sampling strategy for preference elicitation. This method selects queries that are expected to provide the greatest improvement to the utility of the final recommended outcome. We ground our preference model in the probabilistic Plackett-Luce framework for pairwise data. To demonstrate the versatility of our approach, we present two distinct experiments: first, an implementation using matrix factorization for a classic movie recommendation task, and second, an implementation using a neural network for a complex candidate selection scenario in university admissions. Experimental results demonstrate that our framework provides a more accurate, data-efficient, and user-centric paradigm for personalized ranking.

Problem

• Learning preferences from pairwise comparisons to overcome rating biases
• Optimizing task-specific utility functions for personalized recommendations
• Developing active sampling strategy to improve recommendation outcomes

Innovation

• Utility-based active pairwise sampling strategy
• Model-agnostic framework optimizing task-specific utility functions
• Plackett-Luce probabilistic preference modeling from comparisons