An Efficient Large-scale Semi-supervised Multi-label Classifier Capable of Handling Missing labels

TL;DR

This paper introduces a novel non-linear embedding-based multi-label classifier that effectively handles large-scale datasets, missing labels, label correlations, and unlabeled data, outperforming existing methods.

Contribution

The paper presents the first multi-label classifier that simultaneously addresses large-scale data, missing labels, tail label prediction, and unlabeled data exploitation using a non-linear embedding approach.

Findings

Outperforms state-of-the-art classifiers in prediction accuracy.

Reduces training time significantly compared to existing methods.

Effectively predicts infrequent tail labels.

Abstract

Multi-label classification has received considerable interest in recent years. Multi-label classifiers have to address many problems including: handling large-scale datasets with many instances and a large set of labels, compensating missing label assignments in the training set, considering correlations between labels, as well as exploiting unlabeled data to improve prediction performance. To tackle datasets with a large set of labels, embedding-based methods have been proposed which seek to represent the label assignments in a low-dimensional space. Many state-of-the-art embedding-based methods use a linear dimensionality reduction to represent the label assignments in a low-dimensional space. However, by doing so, these methods actually neglect the tail labels - labels that are infrequently assigned to instances. We propose an embedding-based method that non-linearly embeds the label vectors using an stochastic approach, thereby predicting the tail labels more accurately. Moreover, the proposed method have excellent mechanisms for handling missing labels, dealing with large-scale datasets, as well as exploiting unlabeled data. With the best of our knowledge, our proposed method is the first multi-label classifier that simultaneously addresses all of the mentioned challenges. Experiments on real-world datasets show that our method outperforms stateof-the-art multi-label classifiers by a large margin, in terms of prediction performance, as well as training time.