Abstract
Multi-label classification tackles the problems in which each instance is associated with multiple labels. Due to the interdependence among labels, exploiting label correlations is the main means to enhance the performances of classifiers and a variety of corresponding multi-label algorithms have been proposed. Among those algorithms Classifier Chains (CC) is one of the most effective methods. It induces binary classifiers for each label, and these classifiers are linked in a chain. In the chain, the labels predicted by previous classifiers are used as additional features for the current classifier. The original CC has two shortcomings which potentially decrease classification performances: random label ordering, noise in original and additional features. To deal with these problems, we propose a novel and effective algorithm named LSF-CC, i.e. Label Specific Features based Classifier Chain for multi-label classification. At first, a feature estimating technique is employed to produce a list of most relevant features and labels for each label. According to these lists, we define a chain to guarantee that the most frequent labels that appear in these lists are top-ranked. Then, label specific features can be selected from the original feature space and label space. Based on these label specific features, corresponding binary classifiers are learned for each label. Experiments on several multi-label data sets from various domains have shown that the proposed method outperforms well-established approaches.
Highlights
The traditional single-label classification aims to assign a label for each instance from a finite set of labels, including binary and multi-class classification [1]
Because the wrong predictions would propagate along the chain, and an inadequate label ordering can potentially decrease the performance of classifier chain methods. Another important issue of the classifier chain method is that the features, no matter the original feature or the additional features, usually include redundant and irrelevant features, which may bring disadvantages to learning algorithms. To deal with those problems, LSF-Classifier Chains (CC) optimizes the label ordering in a chain of classifiers according to label correlations and selects label specific features from original feature space and label space
In order to obtain insight or deeper understanding of LSF-CC, we explore the performance of LSF-CC which extracts label specific features from original feature spaces rather than label space
Summary
The traditional single-label classification aims to assign a label for each instance from a finite set of labels, including binary and multi-class classification [1]. Another important issue of the classifier chain method is that the features, no matter the original feature or the additional features, usually include redundant and irrelevant features, which may bring disadvantages to learning algorithms To deal with those problems, LSF-CC optimizes the label ordering in a chain of classifiers according to label correlations and selects label specific features from original feature space and label space. The key contributions of our method are summarized as follows: 1) Our proposed algorithm LSF-CC figures out the two shortcomings of traditional classifier chain based algorithms which can potentially decrease their performance: the random label ordering and the noises in original and additional features.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
