Class-Level Feature Selection Method Using Feature Weighted Growing Self-Organising Maps

2025-03-14

AI Summary

Traditional global feature selection methods neglect inter-class discriminability, hindering class-level interpretability analysis. To address this, we propose a class-specific feature selection framework that learns discriminative features independently for each class. Our core method integrates an incremental Growing Self-Organizing Map (GSOM) with a class-conditional feature weighting mechanism—constituting the first such fusion—enabling fine-grained, adaptive modeling of class-wise feature importance. The approach requires no prior specification of the number of classes and supports dynamic topology evolution during learning. Evaluated on standard multi-class benchmark datasets, it consistently outperforms state-of-the-art feature selection algorithms: achieving significant improvements in classification accuracy, delivering transparent, class-specific interpretability, and maintaining low computational overhead.

Technology Category

• Machine Learning: Dimensionality Reduction/Feature Selection
• Search and Optimization: Mixed Discrete/Continuous Search
• Computer Vision: Representation Learning for Vision

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

There have been several attempts to develop Feature Selection (FS) algorithms capable of identifying features that are relevant in a dataset. Although in certain applications the FS algorithms can be seen to be successful, they have similar basic limitations. In all cases, the global feature selection algorithms seek to select features that are relevant and common to all classes of the dataset. This is a major limitation since there could be features that are specifically useful for a particular class while irrelevant for other classes, and full explanation of the relationship at class level therefore cannot be determined. While the inclusion of such features for all classes could cause improved predictive ability for the relevant class, the same features could be problematic for other classes. In this paper, we examine this issue and also develop a class-level feature selection method called the Feature Weighted Growing Self-Organising Map (FWGSOM). The proposed method carries out feature analysis at class level which enhances its ability to identify relevant features for each class. Results from experiments indicate that our method performs better than other methods, gives explainable results at class level, and has a low computational footprint when compared to other methods.

Problem

• Identifies class-specific relevant features in datasets
• Overcomes limitations of global feature selection methods
• Develops FWGSOM for class-level feature analysis

Innovation

• Class-level feature selection using FWGSOM
• Identifies class-specific relevant features
• Low computational footprint, explainable results