Hierarchical Data Augmentation and the Application in Text Classification

Yu, Shujuan; Ye, Jie; Liu, Danlei; Li, Runqi; Zhang, Yun; Zhao, Shengmei

doi:10.1109/access.2019.2960263

Cited by 23 publications

(16 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Other studies have used data noising as smoothing [12] and predictive language models for SR [13]. Hierarchical Data Augmentation (HDA) augments data by extracting the most relevant parts from sentences and connecting them in order using a hierarchical attention network model [14]. Most sentence data augmentation research focuses on one language.…”

Section: Related Workmentioning

confidence: 99%

Iterative Translation-Based Data Augmentation Method for Text Classification Tasks

Lee¹,

Ling

Choi

2021

IEEE Access

View full text Add to dashboard Cite

A well-known limitation of existing rule-based text augmentation is that it cannot be applied to other languages because it depends on grammatical and structural characteristics. Moreover, most text Generative Adversarial Networks (GAN) are unstable in training due to inefficient generator optimization and rely on maximum likelihood pre-training. This paper addresses the above problems by proposing a novel augmentation method with a Sentence Generator (SG) and Sentence Discriminator (SD) for Iterative Translation-based Data Augmentation (ITDA). This paper makes three original contributions. First, the ITDA SG is designed to provide universal multiple-language support by generating comprehensive augmented sentences through serial and parallel iterations of an existing translator, such as Google Translate. Second, given that the quality of the generated sentences varies depending on the translation combination or the type of sentence, the ITDA addresses this issue using a discriminator to achieve sentence augmentation, which can select high-quality augmented data using a text classifier. Third, the ITDA can perform sentence augmentation for 109 different languages using discriminators based on text classifiers trained for a specific language or type of data set. Extensive experiments are conducted to evaluate the efficacy of the ITDA using a Convolutional Neural Network (CNN), Bidirectional Long Short-Term Memory (BiLSTM), CNN-BiLSTM, and self-attention. The results demonstrate that when the ITDA is applied to 480 sentence classification tasks, the average accuracy increases by 4.24%.

show abstract

Section: Related Workmentioning

confidence: 99%

Iterative Translation-Based Data Augmentation Method for Text Classification Tasks

Lee¹,

Ling

Choi

2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Recently, a new proposed method in [21] showed the DA technique-enabled construction of an accurate model with small whole-body CT images. In domains other than images datasets, DA has also been proven effective in, e.g., text classification [22] [23] [24], speech recognition [25] [26], and EEG classification [27]. In this study, we propose a novel method for DA that can be applied to various forms of data for many applications including image and text classification.…”

Section: Related Workmentioning

confidence: 99%

“…The first scenario is the original data without augmentation. The second is partially augmented data without balance via adding more samples only for small categories (22,37). The third scenario is augmented and balanced data.…”

Section: Experimental Settingsmentioning

confidence: 99%

“…Whatever the combination, the performance is not clearly affected. c) The last two classes (22,37): The samples were augmented by adding samples from G, T, and C, so the samples are (66,71). The augmented and balanced data represent a more comprehensive set of possible data points, thus minimizing the distance between the training and validation set as well as any future testing sets, which increases accuracy.…”

Section: Experimental Settingsmentioning

confidence: 99%

See 1 more Smart Citation

Novel Robust Augmentation Approach Based on Sensing Features for Data Classification

Alajmi¹,

Awedat

2021

IEEE Access

View full text Add to dashboard Cite

Learning networks have become extremely powerful tools for data classification. One drawback in many of these applications, however, is that they require a large dataset for efficient performance. Medical imaging, for example, cannot make large enough databases available due to strong data privacy concerns and high data collection costs. Fortunately, data augmentation can provide a solution to improve the classification performance of these systems. This paper proposes a novel method for data augmentation based on the compressive sensing technique. Our approach uses sensing matrices to manipulate data features into different spaces in a compressed form, which are then combined and used as extensions of the original limited dataset. This new feature representation is allowing for a more robust learning network that thereby enables improved data classification. When applying our approach to binary-class and multi-class datasets, results showed that our approach can significantly improve the classifier performance and enhance classification results for both datasets.

show abstract

“…Data augmentation can create several new feature spaces and increase the amount of training data without additional ground truth labels, which has been widely used to improve the performance and generalizability of downstream predictive models. Many works have proposed data augmentation technologies on different types of features, such as images [7,28,15], texts [12,29], vectorized features [18,6], etc. However, how to effectively augment graph data remain a challenging problem, as graph data is more complex and has non-Euclidean structures.…”

Section: Introductionmentioning

confidence: 99%

Data Augmentation for Graph Convolutional Network on Semi-Supervised Classification

Tang

Qiao

Hong

et al. 2021

Preprint

View full text Add to dashboard Cite

Data augmentation aims to generate new and synthetic features from the original data, which can identify a better representation of data and improve the performance and generalizability of downstream tasks. However, data augmentation for graph-based models remains a challenging problem, as graph data is more complex than traditional data, which consists of two features with different properties: graph topology and node attributes. In this paper, we study the problem of graph data augmentation for Graph Convolutional Network (GCN) in the context of improving the node embeddings for semi-supervised node classification. Specifically, we conduct cosine similarity based cross operation on the original features to create new graph features, including new node attributes and new graph topologies, and we combine them as new pairwise inputs for specific GCNs. Then, we propose an attentional integrating model to weighted sum the hidden node embeddings encoded by these GCNs into the final node embeddings. We also conduct a disparity constraint on these hidden node embeddings when training to ensure that non-redundant information is captured from different features. Experimental results on five real-world datasets show that our method improves the classification accuracy with a clear margin(+2.5% -+84.2%) than the original GCN model.

show abstract

Hierarchical Data Augmentation and the Application in Text Classification

Cited by 23 publications

References 15 publications

Iterative Translation-Based Data Augmentation Method for Text Classification Tasks

Iterative Translation-Based Data Augmentation Method for Text Classification Tasks

Novel Robust Augmentation Approach Based on Sensing Features for Data Classification

Data Augmentation for Graph Convolutional Network on Semi-Supervised Classification

Contact Info

Product

Resources

About