Investigating the effect of combining GRU neural networks with handcrafted features for religious hatred detection on Arabic Twitter space

Albadi, Nuha; Kurdi, Maram; Mishra, Shivakant

doi:10.1007/s13278-019-0587-5

Cited by 46 publications

(25 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our previous work [3,4] appears to be the only one focusing on hate speech detection and analysis in Arabic social media. Our study revealed that religious hate speech is widespread on Arabic Twitter.…”

Section: Online Hate Speechmentioning

confidence: 99%

“…In our recent work on hate speech in Arabic social media [3,4], we showed that Arabic Twitter is awash with religious hatred which we defined as "a speech that is insulting, offensive, or hurtful and is intended to incite hate, discrimination, or violence against an individual or a group of people on the basis of religious beliefs or lack thereof". Having such a large volume of hate speech and knowing that ISIS and other radical organizations have been using bots to push their extreme ideologies [6,7], we hypothesize that bots may be to blame for a significant amount of this widespread hatred.…”

Section: Introductionmentioning

confidence: 99%

“…In this work, we build a novel regression model, based on linguistic, content, behavioral and topic features to detect Arabic Twitter bots to understand the impact of bots in spreading religious hatred in Arabic Twitter space. In particular, we quantitatively code and analyze a representative sample of 450 accounts disseminating hate speech from the dataset constructed in our previous work [3,4] for bot-like behavior. We compare our assigned bot-likelihood scores to those of Botometer [18], a well-known machine-learning-based bot detection tool, and we show that Botometer performs a little above average in detecting Arabic bots.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Hateful People or Hateful Bots?

Albadi

Kurdi

Mishra

2019

Proc. ACM Hum.-Comput. Interact.

Self Cite

View full text Add to dashboard Cite

Arabic Twitter space is crawling with bots that fuel political feuds, spread misinformation, and proliferate sectarian rhetoric. While efforts have long existed to analyze and detect English bots, Arabic bot detection and characterization remains largely understudied. In this work, we contribute new insights into the role of bots in spreading religious hatred on Arabic Twitter and introduce a novel regression model that can accurately identify Arabic language bots. Our assessment shows that existing tools that are highly accurate in detecting English bots don't perform as well on Arabic bots. We identify the possible reasons for this poor performance, perform a thorough analysis of linguistic, content, behavioral and network features, and report on the most informative features that distinguish Arabic bots from humans as well as the differences between Arabic and English bots. Our results mark an important step toward understanding the behavior of malicious bots on Arabic Twitter and pave the way for a more effective Arabic bot detection tools.

show abstract

Section: Online Hate Speechmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Hateful People or Hateful Bots?

Albadi

Kurdi

Mishra

2019

Proc. ACM Hum.-Comput. Interact.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Zhang et al [33] introduced a method combining the one-dimensional CNN and the single GRU network, they experimented on the dataset of the Twitter platform and obtained an increase between 1 and 13% in F1-score. Albadi et al [34] tried to combine feature engineering with RNN to detect religious hate speech on Twitter platform, they collected 6000 Arabic posts using Twitter search API, and the experimental result turned out that the single GRU layer with pre-trained word embeddings provided best precision (0.76) and F1-score (0.77), while training the same neural network on additional time, user and content features can provide better recall (0.84).…”

Section: Cyberbullying Detectionmentioning

confidence: 99%

Cyberbullying Detection in Social Networks Using Bi-GRU with Self-Attention Mechanism

et al. 2021

View full text Add to dashboard Cite

With the propagation of cyberbullying in social networks as a trending subject, cyberbullying detection has become a social problem that researchers are concerned about. Developing intelligent models and systems helps detect cyberbullying automatically. This work focuses on text-based cyberbullying detection because it is the commonly used information carrier in social networks and is the widely used feature in this regard studies. Motivated by the documented success of neural networks, we propose a complete model combining the bidirectional gated recurrent unit (Bi-GRU) and the self-attention mechanism. In detail, we introduce the design of a GRU cell and Bi-GRU’s advantage for learning the underlying relationships between words from both directions. Besides, we present the design of the self-attention mechanism and the benefit of this joining for achieving a greater performance of cyberbullying classification tasks. The proposed model could address the limitation of the vanishing and exploding gradient problems. We avoid using oversampling or downsampling on experimental data which could result in the overestimation of evaluation. We conduct a comparative assessment on two commonly used datasets, and the results show that our proposed method outperformed baselines in all evaluation metrics.

show abstract

“…Detecting hate speech on Arabic twittersphere is very promising research tread in literature nowadays. Albadi et al [28] published first publicly available annotated dataset and three lexicons with hate scores in order to detect religious hate speech from Arabic tweets. They classified extracted tweets as hate and not hate speech in terms of lexicon based, n-gram, GRU plus word embedding based, and GRU word embedding with handcrafted features including temporal, user and content features.…”

Section: Related Workmentioning

confidence: 99%

Identification of Abusive Behavior Towards Religious Beliefs and Practices on Social Media Platforms

Ahammad¹,

Uddin²,

Yesmin³

et al. 2021

IJACSA

View full text Add to dashboard Cite

The ubiquitous use of social media has enabled many people, including religious scholars and priests, to share their religious views. Unfortunately, exploiting people's religious beliefs and practices, some extremist groups intentionally or unintentionally spread religious hatred among different communities and thus hamper social stability. This paper aims to propose an abusive behavior detection approach to identify hatred, violence, harassment, and extremist expressions against people of any religious belief on social media. For this, first religious posts from social media users' activities are captured and then the abusive behaviors are identified through a number of sequential processing steps. In the experiment, Twitter has been chosen as an example of social media for collecting dataset of six major religions in English Twittersphere. In order to show the performance of the proposed approach, five classic classifiers on n-gram TF-IDF model have been used. Besides, Long Short-term Memory (LSTM) and Gated Recurrent Unit (GRU) classifiers on trained embedding and pre-trained GloVe word embedding models have been used. The experimental result showed 85% accuracy in terms of precision. However, to the best of our knowledge, this is the first work that will be able to distinguish between hateful and non-hateful contents in other application domains on social media in addition to religious context.

show abstract

Investigating the effect of combining GRU neural networks with handcrafted features for religious hatred detection on Arabic Twitter space

Cited by 46 publications

References 28 publications

Hateful People or Hateful Bots?

Hateful People or Hateful Bots?

Cyberbullying Detection in Social Networks Using Bi-GRU with Self-Attention Mechanism

Identification of Abusive Behavior Towards Religious Beliefs and Practices on Social Media Platforms

Contact Info

Product

Resources

About