Abstract:Unsupervised feature selection is fundamental in statistical pattern recognition, and has drawn persistent attention in the past several decades. Recently, much work has shown that feature selection can be formulated as nonlinear dimensionality reduction with discrete constraints. This line of research emphasizes utilizing the manifold learning techniques, where feature selection and learning can be studied based on the manifold assumption in data distribution. Many existing feature selection methods such as Laplacian score, SPEC (spectrum decomposition of graph Laplacian), TR (trace ratio) criterion, MSFS (multi-cluster feature selection) and EVSC (eigenvalue sensitive criterion) apply the basic properties of graph Laplacian, and select the optimal feature subsets which best preserve the manifold structure defined on the graph Laplacian. In this paper, we propose a new feature selection perspective from locally linear embedding (LLE), which is another popular manifold learning method. The main difficulty of using LLE for feature selection is that its optimization involves quadratic programming and eigenvalue decomposition, both of which are continuous procedures and different from discrete feature selection. We prove that the LLE objective can be decomposed with respect to data dimensionalities in the subset selection problem, which also facilitates constructing better coordinates from data using the principal component analysis (PCA) technique. Based on these results, we propose a novel unsupervised feature selection algorithm, called locally linear selection (LLS), to select a feature subset representing the underlying data manifold. The local relationship among samples is computed from the LLE formulation, which is then used to estimate the contribution of each individual feature to the underlying manifold structure. These contributions, represented as LLS scores, are ranked and selected as the candidate solution to feature selection. We further develop a locally linear rotation-selection (LLRS) algorithm which extends LLS to identify the optimal coordinate subset from a new space. Experimental results on real-world datasets show that our method can be more effective than Laplacian eigenmap based feature selection methods.

FS-MGKC: Feature selection based on structural manifold learning with multi-granularity knowledge coordination

Semi-supervised feature selection based on structure and constraints preserving

Feature Extraction Via Multi-View Non-Negative Matrix Factorization with Local Graph Regularization.

Geodesic Based Semi-supervised Multi-manifold Feature Extraction

Semisupervised Feature Selection via Structured Manifold Learning

Unsupervised feature selection for multi-cluster data

Non-convex feature selection based on feature correlation representation and dual manifold optimization

Joint Adaptive Graph and Structured Sparsity Regularization for Unsupervised Feature Selection

Unsupervised Feature Selection Algorithm Based on Dual Manifold Re-ranking

Fuzzy Neighborhood-Based Manifold Learning and Feature Weight Matrix for Multilabel Feature Selection

Unsupervised feature selection method based on dual manifold learning and dual spatial latent representation

Multi-target feature selection with subspace learning and manifold regularization

Detecting Local Manifold Structure for Unsupervised Feature Selection

Adaptive Unsupervised Feature Selection with Structure Regularization

Semi-Supervised Multiview Feature Selection With Adaptive Graph Learning

Multi-Target Feature Selection with Adaptive Graph Learning and Target Correlations

Dual Sparse Structured Subspaces and Graph Regularisation for Particle Swarm Optimisation-Based Multi-Label Feature Selection

Sparse Feature Selection Via Local Feature and High-Order Label Correlation

Non-negative multi-label feature selection with dynamic graph constraints

Efficient multi-view semi-supervised feature selection

Multi-level regularization-based unsupervised multi-view feature selection with adaptive graph learning