A Structure-Induced Framework for Multi-Label Feature Selection With Highly Incomplete Labels

Abstract

Feature selection has shown significant promise in improving the effectiveness of multilabel learning by constructing a reduced feature space. Previous studies typically assume that label assignment is complete or partially complete; however, missing-label and unlabeled data are commonplace and accompanying occurrences in real applications due to the high expense of manual annotation and label ambiguity. We call this ''highly incomplete labels'' problem. Such label incompleteness severely damages the inherent label structures and masks true label correlations. In this paper, we propose a novel structureinduced feature selection model to simultaneously identify the most discriminative features and recover the highly incomplete labels. To our best knowledge, it is the first attempt to explore the local density structure of data to capture the intricate feature-label dependency in the highly incomplete learning scenarios. Feature selection is guided by the label structure reconstruction, and highly incomplete labels are recovered via the structure transferred from feature space. In this elegant manner, the processes of selecting discriminative features and recovering incomplete labels are coupled in a unified optimization framework. Comprehensive experiments on public benchmark datasets demonstrate the superiority of the proposed approach.

Unified Graph-Based Missing Label Propagation Method for Multilabel Text Classification

“…The feature space is built using the k-highest similarity neighborhoods to maintain the intrinsic local correlation information. To ensure feature space representation validity in recovering label structures, the weight matrix 𝑊 of the feature space similarity is defined as in the following Equation (2) [2,22]:…”

“…Existing methods for handling missing labels in multilabel learning are built based on the assumption that missing label information of an instance can be propagated from its k-nearest neighbors [3,13]. Some of these methods use the first-order label correlation exploitation strategy which ignores label correlations [2][3][4].…”

“…Different from most of the existing algorithms [2,4,13], in multilabel classification with missing labels, this work defines the missing label recovery as a function of both (1) information propagated from its k-nearest neighbors based on the feature space-based similarity and label distribution-based similarity to the label space, and (2) label correlation and label-specific feature learning, which are combined into a single framework. A high-order label correlation is learned from the incomplete training data and applied to augment the missing label matrix and guide the construction of multilabel classification models.…”

“…Handling high dimensionality and feature correlations in multilabel learning may not effectively work if it does not consider the missing label problem (incomplete and noisy label space). Most contemporary approaches treat this problem as a supervised weak-label learning problem, assuming that there are enough partially labeled examples available [2][3][4]. Collecting or annotating such instances, on the other hand, is costly and time consuming.…”

Multi-label feature selection based on fuzzy neighborhood rough sets

“…With the increasing availability of multi-label data related to multiple labels in an instance, a great quantity of feature selection methods for multi-label learning are developed to reduce dimensions and improve learning performance [14][15][16][17]. These methods commonly can be divided into three categories: filter [18][19][20], wrapper [21,22] and embedded [23] methods, where the filter method is independent of the specific learner, and it has less computation cost and stronger generalization ability.…”