Linguistic Knowledge in Data Augmentation for Natural Language Processing: An Example on Chinese Question Matching

TL;DR

This study explores how linguistic knowledge influences data augmentation in Chinese NLP, showing that simple text editing techniques, even with linguistic enhancement, do not significantly outperform basic methods, highlighting the importance of training data quantity.

Contribution

It introduces two data augmentation methods for Chinese question matching and evaluates their effectiveness, revealing limitations of simple text editing techniques regardless of linguistic knowledge integration.

Findings

No significant performance difference between augmented datasets with or without linguistic enhancement.

Simple text editing techniques are limited in producing high-quality paraphrastic data.

Large training datasets are necessary to mitigate false positives from basic augmentation methods.

Abstract

To investigate the role of linguistic knowledge in data augmentation (DA) for Natural Language Processing (NLP), we designed two adapted DA programs and applied them to LCQMC (a Large-scale Chinese Question Matching Corpus) for a binary Chinese question matching classification task. The two DA programs produce augmented texts by five simple text editing operations (or DA techniques), largely irrespective of language generation rules, but one is enhanced with a pre-trained n-gram language model to fuse it with prior linguistic knowledge. We then trained four neural network models (BOW, CNN, LSTM, and GRU) and a pre-trained model (ERNIE-Gram) on the LCQMCs train sets of varying size as well as the related augmented train sets produced by the two DA programs. The results show that there are no significant performance differences between the models trained on the two types of augmented train sets, both when the five DA techniques are applied together or separately. Moreover, due to the inability of the five DA techniques to make strictly paraphrastic augmented texts, the results indicate the need of sufficient amounts of training examples for the classification models trained on them to mediate the negative impact of false matching augmented text pairs and improve performance, a limitation of random text editing perturbations used as a DA approach. Similar results were also obtained for English.