Lifelong Learning of Hate Speech Classification on Social Media

Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 2304–2314 June 6–11, 2021. ©2021 Association for Computational Linguistics 2304 Lifelong Learning of Hate Speech Classification on Social Media Jing Qian†, Hong Wang†, Mai ElSherief∗, Xifeng Yan† † University of California, Santa Barbara ∗ Georgia Institute of Technology {jing_qian,hongwang600,xyan}@cs.ucsb.edu mai.h.sherief@gmail.com Abstract Existing work on automated hate speech clas- sification assumes that the dataset is fixed and the classes are pre-defined. However, the amount of data in social media increases every day, and the hot topics changes rapidly, requir- ing the classifiers to be able to continuously adapt to new data without forgetting the previ- ously learned knowledge. This ability, referred to as lifelong learning, is crucial for the real- word application of hate speech classifiers in social media. In this work, we propose life- long learning of hate speech classification on social media. To alleviate catastrophic forget- ting, we propose to use Variational Represen- tation Learning (VRL) along with a memory module based on LB-SOINN (Load-Balancing Self-Organizing Incremental Neural Network). Experimentally, we show that combining vari- ational representation learning and the LB- SOINN memory module achieves better per- formance than the commonly-used lifelong learning techniques. 1 Introduction With the rapid rise in user-generated web con- tent, the scale and complexity of online hate have reached unprecedented levels in recent years. ADL (Anti-Defamation League) conducted a nationally representative survey of Americans in December 2018 and the report shows that over half (53%) of Americans experienced some type of online ha- rassment.1 This number is higher than the 41% reported to a comparable question asked in 2017 by the Pew Research Center (Center, 2017). To address the growing online hate, a great deal of research has focused on automatic hate speech clas- sification. Most of the previous work focuses on binary classification (Warner and Hirschberg, 2012; Zhong et al., 2016; Nobata et al., 2016; Gao et al., 2017; Qian et al., 2018b) or coarse-grained multi- 1https://www.adl.org/onlineharassment Figure 1: An illustration of our proposed task. hgi: the ith hate group. The model is trained on a sequence of sub-datasets, split by their hate ideologies, e.g., anti- Muslim and Kuklux Klan. The task on each sub-dataset is to identify the hate group given the tweet. class classification (Waseem and Hovy, 2016; Bad- jatiya et al., 2017; Davidson et al., 2017). Qian et al. (2018a) argue that fine-grained classification is necessary for fine-grained hate speech analysis. The Southern Poverty Law Center (SPLC) moni- tors hate groups throughout the United States by a variety of methodologies to determine the activities of groups and individuals, including reviewing hate group publications.2 Therefore, instead of differen- tiating normal posts from the other offensive ones, Qian et al. (2018a) propose a more fine-grained hate speech classification task that attributes hate groups to individual tweets. However, a common limitation of all the research mentioned above is that they assume the dataset to be static and train the classifiers on each isolated …

Content truncated to 500 words