A novel dropout mechanism with label extension schema toward text emotion classification

Zongxi LI, Xianming LI, Haoran XIE, Fu Lee WANG, Mingming LENG, Qing LI, Xiaohui TAO

Research output: Journal PublicationsJournal Article (refereed)peer-review

Abstract

Researchers have been aware that emotion is not one-hot encoded in emotion-relevant classification tasks, and multiple emotions can coexist in a given sentence. Recently, several works have focused on leveraging a distribution label or a grayscale label of emotions in the classification model, which can enhance the one-hot label with additional information, such as the intensity of other emotions and the correlation between emotions. Such an approach has been proven effective in alleviating the overfitting problem and improving the model robustness by introducing a distribution learning component in the objective function. However, the effect of distribution learning cannot be fully unfolded as it can reduce the model’s discriminative ability within similar emotion categories. For example, “Sad” and “Fear” are both negative emotions. To address such a problem, we proposed a novel emotion extension scheme in the prior work (Li, Chen, Xie, Li, and Tao, 2021). The prior work incorporated fine-grained emotion concepts to build an extended label space, where a mapping function between coarse-grained emotion categories and fine-grained emotion concepts was identified. For example, sentences labeled “Joy” can convey various emotions such as enjoy, free, and leisure. The model can further benefit from the extended space by extracting dependency within fine-grained emotions when yielding predictions in the original label space. The prior work has shown that it is more apt to apply distribution learning in the extended label space than in the original space. A novel sparse connection method, i.e., Leaky Dropout, is proposed in this paper to refine the dependency-extraction step, which further improves the classification performance. In addition to the multiclass emotion classification task, we extensively experimented on sentiment analysis and multilabel emotion prediction tasks to investigate the effectiveness and generality of the label extension schema.
Original languageEnglish
Article number103173
JournalInformation Processing and Management
Volume60
Issue number2
Early online date30 Nov 2022
DOIs
Publication statusE-pub ahead of print - 30 Nov 2022

Bibliographical note

This document is the results of the research project funded by a grant from the Research Grants Council of the Hong Kong Special Administrative Region, China (UGC/FDS16/E01/19), and Faculty Research Grants (DB22B4, DB22B7, and DB23A3) and Lam Woo Research Fund (LWP20019) of Lingnan University, Hong Kong, and the General Research Fund from the Hong Kong Research Grants Council (Project no. 11204919).

This article is an extended journal version based on our conference paper, which has been published at Li, Chen, Xie, Li, and Tao (2021). We have sufficient new contents and contributions for this extended journal version according to the journal regulations. Some notations, equations, algorithms, descriptions, tables and figures may be re-used from this paper for smooth presentations.

Publisher Copyright: © 2022 The Author(s)

Keywords

  • Leaky dropout
  • Emotion classification
  • Sentiment analysis
  • Label extension
  • Distribution learning

Fingerprint

Dive into the research topics of 'A novel dropout mechanism with label extension schema toward text emotion classification'. Together they form a unique fingerprint.

Cite this