Mono vs Multilingual BERT for Hate Speech Detection and Text Classification: A Case Study in Marathi

Velankar, Abhishek; Patil, Hrushikesh; Joshi, Raviraj

doi:10.1007/978-3-031-20650-4_10

Cited by 27 publications

(17 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Computer science highlighted that the difficulties in defining OHS affect the process of detecting this content where deep learning, machine learning (Bhawal, et al, 2021; Roy et al, 2020), and annotators play an important role. The use of code-mixed language (Roy et al, 2022 ) and undesired biases (Velankar et al, 2022) make OHS detection challenging. By focussing on context-dependent factors, these contributions failed to consider the moral judgement process affecting human work when annotators manually detect and label OHS (Velankar, et al, 2022).…”

Section: Theoretical Frameworkmentioning

confidence: 99%

“…The use of code-mixed language (Roy et al, 2022 ) and undesired biases (Velankar et al, 2022) make OHS detection challenging. By focussing on context-dependent factors, these contributions failed to consider the moral judgement process affecting human work when annotators manually detect and label OHS (Velankar, et al, 2022).…”

Section: Theoretical Frameworkmentioning

confidence: 99%

“…Literature has addressed OHS from a variety of perspectives, including law (Becker et al, 2000; Boyle, 2001; Citron & Norton, 2011), computer science (Roy, et al, 2020; Velankar, et al, 2022), and policy (Blaya, 2019). For a deeper understanding of the phenomenon and its socio-cultural dimensions, further attention should be devoted to people’s attitudes toward the moral dimension of OHS, including understandings of what constitutes morally acceptable behaviours.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Online Hate Speech as a Moral Issue: Exploring Moral Reasoning of Young Italian Users on Social Network Sites

Ieracitano

Balenzano²,

Girardi³

et al. 2023

Social Science Computer Review

View full text Add to dashboard Cite

Taking a neo-Kohlbergian approach, we explore the moral reasoning of 486 young Italian users of social network sites exposed to moral dilemmas concerning online hate speech. The aims are to understand what moral reasoning schemas they use as they face homophobic, racist, or sexist online hate speech, and what influence personal values and moral disengagement might have on their moral reasoning process. The results reveal the prevalence of Maintaining Norms reasoning (conformity to rules and authority) in making moral decisions concerning online hate speech and confirms the mediating role of Hate Speech Moral Disengagement in the relationship between personal values and the moral reasoning process.

show abstract

Section: Theoretical Frameworkmentioning

confidence: 99%

Section: Theoretical Frameworkmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Online Hate Speech as a Moral Issue: Exploring Moral Reasoning of Young Italian Users on Social Network Sites

Ieracitano

Balenzano²,

Girardi³

et al. 2023

Social Science Computer Review

View full text Add to dashboard Cite

show abstract

“…A study shows that monolingual versions outperform the traditional multilingual models for all datasets. Moreover, better sentence representations are also generated by the monolingual models [43]. However, training and maintenance for each language in a monolingual model is not cost-effective in terms of time and resources.…”

Section: Literature Reviewmentioning

confidence: 99%

A Multi-Layer Holistic Approach for Cursive Text Recognition

et al. 2022

View full text Add to dashboard Cite

Urdu is a widely spoken and narrated language in several South-Asian countries and communities worldwide. It is relatively hard to recognize Urdu text compared to other languages due to its cursive writing style. The Urdu text script belongs to a non-Latin cursive family script like Arabic, Hindi and Chinese. Urdu is written in several writing styles, among which `Nastaleeq’ is the most popular and widely used font style. A gap still poses a challenge for localization/detection and recognition of Urdu Nastaleeq text as it follows modified version of Arabic script. This research study presents a methodology to recognize and classify Urdu text in Nastaleeq font, regardless of the text position in the image. The proposed solution is comprised of a two-step methodology. In the first step, text detection is performed using the Connected Component Analysis (CCA) and Long Short-Term Memory Neural Network (LSTM). In the second step, a hybrid Convolution Neural Network and Recurrent Neural Network (CNN-RNN) architecture is deployed to recognize the detected text. The image containing Urdu text is binarized and segmented to produce a single-line text image fed to the hybrid CNN-RNN model, which recognizes the text and saves it in a text file. The proposed technique outperforms the existing ones by achieving an overall accuracy of 97.47%.

show abstract

“…Multilingual versions such as mBERT and XLM-RoBERTA have since been released. Developers have also released monolingual BERT models, such as German BERT and MahaBERT for Marathi (Velankar, Patil, & Joshi, 2022). However, the multilingual versions are still widely used in literature.…”

Section: Related Workmentioning

confidence: 99%

Combating Hate: How Multilingual Transformers Can Help Detect Topical Hate Speech

Srikissoon,

Marivate

EPiC Series in Computing

View full text Add to dashboard Cite

Automated hate speech detection is important to protecting people’s dignity, online experiences, and physical safety in Society 5.0. Transformers are sophisticated pre- trained language models that can be fine-tuned for multilingual hate speech detection. Many studies consider this application as a binary classification problem. Additionally, research on topical hate speech detection use target-specific datasets containing assertions about a particular group. In this paper we investigate multi-class hate speech detection using target-generic datasets. We assess the performance of mBERT and XLM-RoBERTA on high and low resource languages, with limited sample sizes and class imbalance. We find that our fine-tuned mBERT models are performant in detecting gender-targeted hate speech. Our Urdu classifier produces a 31% lift on the baseline model. We also present a pipeline for processing multilingual datasets for multi-class hate speech detection. Our approach could be used in future works on topically focused hate speech detection for other low resource languages, particularly African languages which remain under-explored in this domain.

show abstract

Mono vs Multilingual BERT for Hate Speech Detection and Text Classification: A Case Study in Marathi

Cited by 27 publications

References 11 publications

Online Hate Speech as a Moral Issue: Exploring Moral Reasoning of Young Italian Users on Social Network Sites

Online Hate Speech as a Moral Issue: Exploring Moral Reasoning of Young Italian Users on Social Network Sites

A Multi-Layer Holistic Approach for Cursive Text Recognition

Combating Hate: How Multilingual Transformers Can Help Detect Topical Hate Speech

Contact Info

Product

Resources

About