Algorithm Selection for Recommender Systems via Meta-Learning on Algorithm Characteristics

2025-08-06

AI Summary

Algorithm selection in recommender systems remains challenging: conventional meta-learning treats algorithms as discrete categories, neglecting their intrinsic characteristics. This paper proposes the first personalized algorithm selection framework that jointly models user meta-features and static source-code features of recommendation algorithms, thereby overcoming the limitations of categorical modeling. By automatically extracting code-level metrics—such as computational complexity and call depth—from open-source algorithm implementations, we construct fine-grained algorithm representations. These are fused with user profiles and fed into a supervised performance prediction model for accurate, instance-level recommendation quality estimation. Evaluated on six benchmark datasets, our approach achieves an average NDCG@10 of 0.147—improving by 8.83% over user-feature-only baselines, outperforming the single best-performing algorithm, and reducing the gap to the theoretical optimal selector’s performance by 10.5%. This work advances interpretable and generalizable algorithm selection research in recommendation.

Technology Category

• Search and Optimization: Algorithm Configuration
• Machine Learning: Learning Preferences or Rankings
• Data Mining & Knowledge Management: Recommender Systems

Application Category

• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

The Algorithm Selection Problem for recommender systems-choosing the best algorithm for a given user or context-remains a significant challenge. Traditional meta-learning approaches often treat algorithms as categorical choices, ignoring their intrinsic properties. Recent work has shown that explicitly characterizing algorithms with features can improve model performance in other domains. Building on this, we propose a per-user meta-learning approach for recommender system selection that leverages both user meta-features and automatically extracted algorithm features from source code. Our preliminary results, averaged over six diverse datasets, show that augmenting a meta-learner with algorithm features improves its average NDCG@10 performance by 8.83% from 0.135 (user features only) to 0.147. This enhanced model outperforms the Single Best Algorithm baseline (0.131) and successfully closes 10.5% of the performance gap to a theoretical oracle selector. These findings show that even static source code metrics provide a valuable predictive signal, presenting a promising direction for building more robust and intelligent recommender systems.

Problem

• Selecting best recommender algorithm for users using meta-learning
• Improving algorithm selection by incorporating algorithm features
• Enhancing recommender performance with static source code metrics

Innovation

• Meta-learning with user and algorithm features
• Automated algorithm feature extraction from source code
• Static source code metrics for predictive signal