Text Classification through Glyph-aware Disentangled Character Embedding and Semantic Sub-character Augmentation

TL;DR

This paper introduces a novel glyph-aware disentangled character embedding and semantic sub-character augmentation framework for non-alphabetic language text classification, improving interpretability and performance.

Contribution

It presents a new variational character encoder with glyph-aware embeddings and a semantic augmentation method, enhancing interpretability and classification accuracy.

Findings

GDCE provides interpretable, dimensionally independent embeddings

SSA improves classification performance

Framework achieves competitive results with state-of-the-art models

Abstract

We propose a new character-based text classification framework for non-alphabetic languages, such as Chinese and Japanese. Our framework consists of a variational character encoder (VCE) and character-level text classifier. The VCE is composed of a $\beta$-variational auto-encoder ($\beta$-VAE) that learns the proposed glyph-aware disentangled character embedding (GDCE). Since our GDCE provides zero-mean unit-variance character embeddings that are dimensionally independent, it is applicable for our interpretable data augmentation, namely, semantic sub-character augmentation (SSA). In this paper, we evaluated our framework using Japanese text classification tasks at the document- and sentence-level. We confirmed that our GDCE and SSA not only provided embedding interpretability but also improved the classification performance. Our proposal achieved a competitive result to the state-of-the-art model while also providing model interpretability. Our code is available on https://github.com/IyatomiLab/GDCE-SSA