Dynamic Label Propagation for Semi-supervised Multi-class Multi-label Classification

Abstract

In graph-based semi-supervised learning approaches, the classification rate is highly dependent on the size of the availabel labeled data, as well as the accuracy of the similarity measures. Here, we propose a semi-supervised multi-class/multi-label classification scheme, dynamic label propagation (DLP), which performs transductive learning through propagation in a dynamic process. Existing semi-supervised classification methods often have difficulty in dealing with multi-class/multi-label problems due to the lack in consideration of label correlation; our algorithm instead emphasizes dynamic metric fusion with label information. Significant improvement over the state-of-the-art methods is observed on benchmark datasets for both multiclass and multi-label tasks.

Semi-supervised dictionary learning with label propagation for image classification

“…[28] to construct the weight matrix, our weight matrix is constructed from the reconstruction errors of the unlabeled samples over all classes rather than the distances between any two samples. Intuitively, since sub-dictionary D i is good at representng the ith-class but is poor at representing other classes, any pair of samples is likely to belong to the same class if they achieve minimum reconstruction error in the same class.…”

“…In recent years, semi-supervised learning methods have been widely studied [27][28][29][30][31]. One classical semi-supervised learning method is co-training [29] which utilizes multi-view features to retrain the classifiers to obtain better performance.…”

“…A representative method based on graphs is label propagation (LP), which has been widely used in image classification and ranking. Label propagation algorithms [27,28,[33][34][35] perform class estimation of unlabeled samples by propagating label information from labeled data to unlabeled data. This is done by constructing a weight matrix (or affinity matrix) based on the distance between any two samples.…”

Multi-Label Classification Based on Low Rank Representation for Image Annotation

“…We conduct experiments on Land Cover and five other publicly-available multi-label image datasets, Flags [53], Scene [13], Corel5k [54], MIRFlickr [55] and ESPGame [56], to validate the performance of our proposed MLC-LRR and compare it with five representative and related graph-based multi-label classifiers: MSC [20], TMC [27], FCML [29], Tram [19] and DLP [31]. MSC is a supervised multi-label classifier, and the other four are semi-supervised multi-label classifiers.…”

“…It formulates the transductive multi-label classification as an optimization problem of estimating label composition and provides a closed-form solution. Wang et al [31] proposed a Dynamic Label Propagation (DLP) to infer labels of unlabeled images by using an l 2 -graph, which is constructed by exploiting neighborhood images of an image and neighborhood images of its reciprocal neighbors. However, all of these aforementioned methods for graph construction do not adequately exploit the global structure of data.…”