Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

TRUST-FS: Tensorized Reliable Unsupervised Multi-View Feature Selection for Incomplete Data

Minghui Lu, Yanyong Huang, Minbo Ma, Jinyuan Chang, Dongjie Wang, Xiuwen Yi, Tianrui Li

TL;DR

TRUST-FS tackles incomplete multi-view unsupervised feature selection with missing variables by unifying feature selection, missing-variable imputation, and view-weight learning within a tensorized framework. It introduces Adaptive-Weighted CP decomposition to jointly learn low-dimensional representations and impute data, while a Subjective Logic–driven reliability mechanism refines cross-view similarity graphs to guide selection. The approach yields a reliable, integrated objective that outperforms state-of-the-art baselines on eight datasets across varying missing-data scenarios, with convergence guarantees and manageable complexity. This yields robust feature selection for high-dimensional, multi-view data in real-world incomplete settings, enhancing downstream clustering and interpretability.

Abstract

Multi-view unsupervised feature selection (MUFS), which selects informative features from multi-view unlabeled data, has attracted increasing research interest in recent years. Although great efforts have been devoted to MUFS, several challenges remain: 1) existing methods for incomplete multi-view data are limited to handling missing views and are unable to address the more general scenario of missing variables, where some features have missing values in certain views; 2) most methods address incomplete data by first imputing missing values and then performing feature selection, treating these two processes independently and overlooking their interactions; 3) missing data can result in an inaccurate similarity graph, which reduces the performance of feature selection. To solve this dilemma, we propose a novel MUFS method for incomplete multi-view data with missing variables, termed Tensorized Reliable UnSupervised mulTi-view Feature Selection (TRUST-FS). TRUST-FS introduces a new adaptive-weighted CP decomposition that simultaneously performs feature selection, missing-variable imputation, and view weight learning within a unified tensor factorization framework. By utilizing Subjective Logic to acquire trustworthy cross-view similarity information, TRUST-FS facilitates learning a reliable similarity graph, which subsequently guides feature selection and imputation. Comprehensive experimental results demonstrate the effectiveness and superiority of our method over state-of-the-art methods.

TRUST-FS: Tensorized Reliable Unsupervised Multi-View Feature Selection for Incomplete Data

TL;DR

TRUST-FS tackles incomplete multi-view unsupervised feature selection with missing variables by unifying feature selection, missing-variable imputation, and view-weight learning within a tensorized framework. It introduces Adaptive-Weighted CP decomposition to jointly learn low-dimensional representations and impute data, while a Subjective Logic–driven reliability mechanism refines cross-view similarity graphs to guide selection. The approach yields a reliable, integrated objective that outperforms state-of-the-art baselines on eight datasets across varying missing-data scenarios, with convergence guarantees and manageable complexity. This yields robust feature selection for high-dimensional, multi-view data in real-world incomplete settings, enhancing downstream clustering and interpretability.

Abstract

Multi-view unsupervised feature selection (MUFS), which selects informative features from multi-view unlabeled data, has attracted increasing research interest in recent years. Although great efforts have been devoted to MUFS, several challenges remain: 1) existing methods for incomplete multi-view data are limited to handling missing views and are unable to address the more general scenario of missing variables, where some features have missing values in certain views; 2) most methods address incomplete data by first imputing missing values and then performing feature selection, treating these two processes independently and overlooking their interactions; 3) missing data can result in an inaccurate similarity graph, which reduces the performance of feature selection. To solve this dilemma, we propose a novel MUFS method for incomplete multi-view data with missing variables, termed Tensorized Reliable UnSupervised mulTi-view Feature Selection (TRUST-FS). TRUST-FS introduces a new adaptive-weighted CP decomposition that simultaneously performs feature selection, missing-variable imputation, and view weight learning within a unified tensor factorization framework. By utilizing Subjective Logic to acquire trustworthy cross-view similarity information, TRUST-FS facilitates learning a reliable similarity graph, which subsequently guides feature selection and imputation. Comprehensive experimental results demonstrate the effectiveness and superiority of our method over state-of-the-art methods.

Paper Structure

This paper contains 20 sections, 3 theorems, 36 equations, 9 figures, 3 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Work
  3. The Propose Method
  4. Notations and Problem Definition
  5. Tensorized Joint Learning for IMUFS
  6. Reliable Complementary Graph Learning
  7. Optimization and Analyses
  8. Convergence and Complexity Analysis
  9. Convergence Analysis
  10. Complexity Analysis
  11. Experiment
  12. Experimental Schemes
  13. datasets
  14. Evaluation Metrics
  15. Compared Methods
  16. ...and 5 more sections

Key Result

Theorem 1

The update rules in Algorithm 1 guarantee that the objective function $\mathcal{J}(\hat{\bm{X}}^{(v)}, \bm{W}^{(v)}, \bm{S}^{(v)}, \bm{A}, \bm{P}, \bm{\hat{b}}_v, \bm{H}, \bm{\omega})$ decreases monotonically with each iteration until convergence.

Figures (9)

  • Figure 1: The framework of the proposed TRUST-FS.
  • Figure 2: ACC of different methods on eight datasets under different feature selection ratios.
  • Figure 3: NMI of different methods on eight datasets under different feature selection ratios.
  • Figure 4: ACC of different methods on eight datasets with different missing ratios.
  • Figure 5: NMI of different methods on eight datasets with different missing ratios.
  • ...and 4 more figures

Theorems & Definitions (5)

  • Theorem 1
  • proof
  • Definition 1
  • Lemma 1
  • Lemma 2