FeatureCuts: Feature Selection for Large Data by Optimizing the Cutoff

2025-07-31

AI Summary

To address the low efficiency and high computational overhead of feature selection in large-scale data scenarios, this paper proposes FeatureCuts—a novel feature selection method integrating filter-based ranking with an adaptive dynamic truncation mechanism. Its core innovation lies in an information-preserving optimal truncation point optimization strategy, which automatically determines per-feature retention thresholds to achieve substantial feature compression while preserving model performance. FeatureCuts is designed to synergize with wrapper algorithms (e.g., Particle Swarm Optimization, PSO) for joint optimization. Extensive experiments on multiple public and industrial datasets demonstrate that FeatureCuts reduces the number of features by 15% on average and accelerates computation by up to 99.6%. When combined with PSO, the feature reduction rate increases to 25%, and computational cost decreases further by 66%.

Technology Category

• Machine Learning: Dimensionality Reduction/Feature Selection
• Search and Optimization: Evolutionary Computation
• Data Mining & Knowledge Management: Data Compression

Application Category

• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

In machine learning, the process of feature selection involves finding a reduced subset of features that captures most of the information required to train an accurate and efficient model. This work presents FeatureCuts, a novel feature selection algorithm that adaptively selects the optimal feature cutoff after performing filter ranking. Evaluated on 14 publicly available datasets and one industry dataset, FeatureCuts achieved, on average, 15 percentage points more feature reduction and up to 99.6% less computation time while maintaining model performance, compared to existing state-of-the-art methods. When the selected features are used in a wrapper method such as Particle Swarm Optimization (PSO), it enables 25 percentage points more feature reduction, requires 66% less computation time, and maintains model performance when compared to PSO alone. The minimal overhead of FeatureCuts makes it scalable for large datasets typically seen in enterprise applications.

Problem

• Optimizing feature selection cutoff for large datasets
• Reducing computation time while maintaining model performance
• Enhancing scalability for enterprise-level machine learning

Innovation

• Adaptive feature cutoff selection
• Efficient filter ranking optimization
• Scalable for large enterprise datasets