Word-Class Embeddings for Multiclass Text Classification

TL;DR

This paper introduces supervised word-class embeddings that, when combined with pre-trained embeddings, improve deep learning models' accuracy in multiclass text classification across various datasets and architectures.

Contribution

The paper proposes supervised word-class embeddings (WCEs) that enhance pre-trained embeddings, significantly improving multiclass text classification performance.

Findings

WCEs improve classification accuracy across four neural architectures.

WCEs show consistent gains on six public datasets.

Code implementation is publicly available.

Abstract

Pre-trained word embeddings encode general word semantics and lexical regularities of natural language, and have proven useful across many NLP tasks, including word sense disambiguation, machine translation, and sentiment analysis, to name a few. In supervised tasks such as multiclass text classification (the focus of this article) it seems appealing to enhance word representations with ad-hoc embeddings that encode task-specific information. We propose (supervised) word-class embeddings (WCEs), and show that, when concatenated to (unsupervised) pre-trained word embeddings, they substantially facilitate the training of deep-learning models in multiclass classification by topic. We show empirical evidence that WCEs yield a consistent improvement in multiclass classification accuracy, using four popular neural architectures and six widely used and publicly available datasets for multiclass text classification. Our code that implements WCEs is publicly available at https://github.com/AlexMoreo/word-class-embeddings