Data Augmentation for Multiclass Utterance Classification – A Systematic Study

Abstract

Utterance classification is a key component in many conversational systems. However, classifying real-world user utterances is challenging, as people may express their ideas and thoughts in manifold ways, and the amount of training data for some categories may be fairly limited, resulting in imbalanced data distributions. To alleviate these issues, we conduct a comprehensive survey regarding data augmentation approaches for text classification, including simple random resampling, word-level transformations, and neural text generation to cope with imbalanced data. Our experiments focus on multi-class datasets with a large number of data samples, which has not been systematically studied in previous work. The results show that the effectiveness of different data augmentation schemes depends on the nature of the dataset under consideration.

A Survey on Legal Judgment Prediction: Datasets, Metrics, Models and Challenges

“…Few-shot learning has become a hot spot [87,51,88,89,90] since current works focus more on high-frequency judgment results than few-shot judgment results to ensure enough training data.…”

Data Augmentation Approaches in Natural Language Processing: A Survey

“…Text Structure classification generation prediction Paraphrasing Thesauruses [5], [93], [49], [7], [42], [60], [44], [45], [98] - [42], [43] Embeddings [8], [49] --MLMs [10], [51], [54] [55] -Rules [10], [7], [11] -[99] MT [42], [60], [10], [12], [59], [61], [63], [7], [19], [66], [100], [98] [13], [58] [42], [57], [15] Seq2Seq [18], [68], [101] [18], [102] [18], [16], [67], [17], [103], [82] Noising Swapping [93], [60], [44], [61], [20], [19] -…”