EDA: Easy Data Augmentation Techniques for Boosting Performance on Text Classification Tasks

By Jason Wei, Kai Zou. 2019

2,358 citations recorded by the source corpus.

We present EDA: easy data augmentation techniques for boosting performance on text classification tasks. EDA consists of four simple but powerful operations: synonym replacement, random insertion, random swap, and random deletion. On five text classification tasks, we show that EDA improves performance for both convolutional and recurrent neural networks. EDA demonstrates particularly strong results for smaller datasets; on average, across five datasets, training with EDA while using only 50% of the available training set achieved the same accuracy as normal training with all available data. We also performed extensive ablation studies and suggest parameters for practical use.

Subfields: Artificial Intelligence, Natural Language Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Natural Language Processing

Citations in the corpus

A Survey of Data Augmentation Approaches for NLP (2021, methodology): “but usually offer incremental performance improvements (Li et al., 2017; Wei and Zou, 2019; Wei et al., 2021b).”
Do Not Have Enough Data? Deep Learning to the Rescue! (2019, background): “Data augmentation (Wong et al. 2016) is a common strategy for handling scarce data situations.”
Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time (2022, background): “data augmentation (Wei and Zou, 2019; Ma, 2019)) could lead to more diverse models and better soups.”
Evaluating the Factual Consistency of Abstractive Text Summarization (2019, background): “Wei and Zou (2019) proposed an augmentation framework aimed at boosting performance of text classification models.”
MixText: Linguistically-Informed Interpolation of Hidden Space for Semi-Supervised Text Classification (2020, background): “Recently, Wei and Zou (2019) utilized synonym replacement, random insertion, random swap and random deletion for text data augmentation.”
GraDA: Graph Generative Data Augmentation for Commonsense Reasoning (2022, methodology): “To evaluate this, we use the TextFooler system (Jin et al., 2020; Yang et al., 2020; Wei and Zou, 2019) to generate adversarial text by computing word importance ranking and replacing the most influential words with their synonym in the vector space.”
Towards Zero-Label Language Learning (2021, methodology): “Data augmentation has traditionally been a popular technique for NLP model quality improvement, especially in low-resource regimes (Yu et al., 2018; Wei and Zou, 2019) While traditionally simple heuristics like token-level modiﬁcation has been applied to diversify training samples, more recently…”
Data Augmentation using Pre-trained Transformer Models (2020, background): “In particular, Wei and Zou (2019) show that simple word replacement us-ing knowledge bases like WordNet (Miller, 1998) improves classiﬁcation performance.”
Generalizing from a Few Examples (2019, background): “A recent attempt on using data augmentation for text is reported in [144].”
Supervision Exists Everywhere: A Data Efficient Contrastive Language-Image Pre-training Paradigm (2021, methodology): “Text augmentations We use the EDA (Wei & Zou, 2019) method as our text augmentation strategy, which contains three types of text augmentation strategies: synonym replacement, random swap, and random deletion.”
G-DAug: Generative Data Augmentation for Commonsense Reasoning (2020, methodology): “In traditional data augmentation settings, augmented data are usually mixed with the original training set and are trained together with the original data (Wei and Zou, 2019; Kafle et al., 2017).”
Enhanced Offensive Language Detection Through Data Augmentation (2020, methodology): “Instead of simple word-level or phrase-level replacement(Kobayashi 2018; Wei and Zou 2019), our method provides sentence-level text data augmentation, enriching semantic and syntactic features of our generated texts.”
Multilingual Sentiment Analysis and Toxicity Detection for Text Messages in Russian (2021, methodology): “forming text classification tasks [21].”
Five sources of bias in natural language processing (2021, background): “Wei and Zou (2019) explore data augmentation techniques that improve performance on various text classification tasks.”
SSMBA: Self-Supervised Manifold Based Data Augmentation for Improving Out-of-Domain Robustness (2020, methodology): “, 2017) and word-level methods (Zhang et al., 2015; Xie et al., 2017; Wei and Zou, 2019) are based on human-designed heuristics.”
A Comprehensive Survey of Few-shot Learning: Evolution, Applications, Challenges, and Opportunities (2022, background): “Furthermore, [52] and [53] combine these simple sub-strategies named auto data augmentation.”
Augmenting NLP models using Latent Feature Interpolations (2020, methodology): “In general, it is difficult to come up with rules for language transformation similar to image transformations; hence universal data augmentation techniques have not thoroughly been explored in NLP yet (Wei and Zou, 2019).”