Annotation Framework for Hate Speech Identification in Tweets: Case Study of Tweets During Kenyan Elections

Ombui, Edward; Karani, Moses; Muchemi, Lawrence

doi:10.23919/istafrica.2019.8764868

Cited by 6 publications

(3 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These call for approaches that can adapt to newly seen content out of the original training corpus. Annotating such content is non-trivial and may require substantial time and effort (Poletto et al, 2019;Ombui et al, 2019). Thus, Unsupervised Domain Adaptation (UDA) methods that can adapt without the target domain labels (Ramponi and Plank, 2020), turn out to be attractive in this task.…”

Section: Introductionmentioning

confidence: 99%

Unsupervised Domain Adaptation in Cross-corpora Abusive Language Detection

Bose¹,

Illina²,

Fohr³

2021

Proceedings of the Ninth International Workshop on Natural Language Processing for Social Media

View full text Add to dashboard Cite

The state-of-the-art abusive language detection models report great in-corpus performance, but underperform when evaluated on abusive comments that differ from the training scenario. As human annotation involves substantial time and effort, models that can adapt to newly collected comments can prove to be useful. In this paper, we investigate the effectiveness of several Unsupervised Domain Adaptation (UDA) approaches for the task of cross-corpora abusive language detection. In comparison, we adapt a variant of the BERT model, trained on large-scale abusive comments, using Masked Language Model (MLM) fine-tuning. Our evaluation shows that the UDA approaches result in sub-optimal performance, while the MLM fine-tuning does better in the cross-corpora setting. Detailed analysis reveals the limitations of the UDA approaches and emphasizes the need to build efficient adaptation methods for this task.

show abstract

Section: Introductionmentioning

confidence: 99%

Unsupervised Domain Adaptation in Cross-corpora Abusive Language Detection

Bose¹,

Illina²,

Fohr³

2021

Proceedings of the Ninth International Workshop on Natural Language Processing for Social Media

View full text Add to dashboard Cite

show abstract

“…The review of the literature indicates a growing number of studies that build and manually annotate corpora for offensive language, sentiment analysis, and hate speech identification. The corpora in most of these studies are in English [6,7,8] and in European languages like Portuguese [9,10], German [12], Spanish [13], and Italian [14]. A few studies have built new datasets in Hindi [15], Arabic [16] [17], and Amharic [18].…”

Section: Related Workmentioning

confidence: 99%

Building and Annotating a Codeswitched Hate Speech Corpora

Ombui¹,

Muchemi²,

Wagacha³

2021

IJITCS

Self Cite

View full text Add to dashboard Cite

Presidential campaign periods are a major trigger event for hate speech on social media in almost every country. A systematic review of previous studies indicates inadequate publicly available annotated datasets and hardly any evidence of theoretical underpinning for the annotation schemes used for hate speech identification. This situation stifles the development of empirically useful data for research, especially in supervised machine learning. This paper describes the methodology that was used to develop a multidimensional hate speech framework based on the duplex theory of hate [1] components that include distance, passion, commitment to hate, and hate as a story. Subsequently, an annotation scheme based on the framework was used to annotate a random sample of ~51k tweets from ~400k tweets that were collected during the August and October 2017 presidential campaign period in Kenya. This resulted in a goldstandard codeswitched dataset that could be used for comparative and empirical studies in supervised machine learning. The resulting classifiers trained on this dataset could be used to provide real-time monitoring of hate speech spikes on social media and inform data-driven decision-making by relevant security agencies in government.

show abstract

Section: Introductionmentioning

confidence: 99%

Proceedings of the Ninth International Workshop on Natural Language Processing for Social Media

2021

View full text Add to dashboard Cite

The Moral Foundation Theory suggests five moral foundations that can capture the view of a user on a particular issue. It is widely used to identify sentence-level sentiment. In this paper, we study the nuanced stances and partisan sentiment towards entities of US politicians using Moral Foundation Theory, on two politically divisive issues -Gun Control and Immigration. We define the nuanced stances of the US politicians on these two topics by the grades given by related organizations to the politicians. To conduct this study, we first filter out 74k and 87k tweets on the topics Gun Control and Immigration, respectively, from an existing tweet corpus authored by US parliament members. Then, we identify moral foundations in these tweets using deep relational learning. Finally, doing qualitative and quantitative evaluations on this dataset, we found out that there is a strong correlation between moral foundation usage and politicians' nuanced stances on a particular topic. We also found notable differences in moral foundation usage by different political parties when they address different entities. c : SameIdeology(t1, t2) ∧ SameTopic(t1, t2)∧ SameTime(t1, t2) ∧ HasMF(t1, m) ⇒ HasMF(t2, m) 10 Rujun Han, Qiang Ning, and Nanyun Peng. 2019. Joint event and temporal relation extraction with shared representations and structured prediction. In

show abstract

Annotation Framework for Hate Speech Identification in Tweets: Case Study of Tweets During Kenyan Elections

Cited by 6 publications

References 12 publications

Unsupervised Domain Adaptation in Cross-corpora Abusive Language Detection

Unsupervised Domain Adaptation in Cross-corpora Abusive Language Detection

Building and Annotating a Codeswitched Hate Speech Corpora

Proceedings of the Ninth International Workshop on Natural Language Processing for Social Media

Contact Info

Product

Resources

About