Deep learning for online harassment detection in tweets
The DOI matches its Crossref record; 20 of the 22 references it cites were found in a registry or an index.
Abstract
In recent years, the strong rise in the use of social network platforms such as Twitter has resulted in millions of users sharing their thoughts and opinions about different aspects and events on the micro-blogging platforms. However, the unfiltered exchange of the content and the missing protection of private information can lead to sending or broadcasting bully and harassment messages in different social media platforms. This kind of messages may have serious negative impact on a person such as causing low self-esteem and depression. Therefore, automatic detection of such messages is more than necessary. This paper casts the problem as a classification task and investigates the effectiveness of deep learning to detect online harassment in Large Human-Labeled corpus specially designed for harassment research purpose. To this end, models are considered namely Long short-term memory(LSTM) ,Bidirectional Long Short-Term Memory (BLSTM),Convolutional neural Network (CNN), and compared with other classification models. Obtained results are very encouraging.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
Efficient Estimation of Word Representations in Vector Space
Mikolov, T., Chen, K., Corrado, G., & Dean, J. (2013) · arXiv · doi:10.48550/arxiv.1301.3781 · Cited by 26,568 · cited 2×
“our work address the issue of online harassment detection in large corpus of tweets [12] through the use of: - in one side, different traditional machine learning methods such as : logistic regression, support vector machine,naive Bayes and random forest with two word representations namely character n-gram and bag of…”
MethodConvolutional Neural Networks for Sentence Classification
Kim, Y. (2014) · Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.3115/v1/d14-1181 · Cited by 14,831 · cited 1×
“• The neural architecture :We adopted the same architecture of CNN,LSTM,BLSTM cited in [17] , [18], [19]”
MethodA unified architecture for natural language processing: deep neural networks with multitask learning
Collobert, R., & Weston, J. (2008) · Proceedings of the 25th international conference on Machine learning - ICML '08 · doi:10.1145/1390156.1390177 · Cited by 5,058 · cited 1×
“However, Collobert and Weston were arguably the first to demonstrate the power of pre-trained word embeddings in [11] ], in which they establish word embeddings as a highly effective tool when used in different NLP tasks.”
A Neural Probabilistic Language Model
Bengio, Y., Ducharme, R., Vincent, P., Jauvin, C., Jaz, K., Hofmann, T., Poggio, T., & Shawe-Taylor, J. S. (2000) · Cited by 3,479
Deep learning for computer vision: A brief review
Voulodimos, A., Doulamis, N., Doulamis, A., & Protopapadakis, E. (2018) · Computational Intelligence and Neuroscience · doi:10.1155/2018/7068349 · Cited by 3,403
A neural probabilistic language model
BengioYoshua, DucharmeRéjean, VincentPascal, & JanvinChristian (2003) · Journal of Machine Learning Research · doi:10.5555/944919.944966 · Cited by 2,644
Deep Speech 2: End-to-End Speech Recognition in English and Mandarin
Amodei, D., Anubhai, R., Battenberg, E., Case, C. T., Casper, J., Catanzaro, B., Chen, J., Chrzanowski, M., Coates, A., Diamos, G., et al. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1512.02595 · Cited by 2,110 · cited 1×
“Deep neural networks have contributed to outstanding advancements in many tasks like computer vision [14] and speech recognition [15].”
BackgroundLearning Sentiment-Specific Word Embedding for Twitter Sentiment Classification
Tang, D., Wei, F., Yang, N., Zhou, M., Liu, T., & Qin, B. (2014) · Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.3115/v1/p14-1146 · Cited by 1,181 · cited 1×
“Moreover, SSWE pre-trained on 10 M tweets [13].”
MethodDeep Learning for Hate Speech Detection in Tweets
Badjatiya, P., Gupta, S., Gupta, M., & Varma, V. (2017) · Proceedings of the 26th International Conference on World Wide Web Companion - WWW '17 Companion · doi:10.1145/3041021.3054223 · Cited by 849 · cited 1×
“[8] used different architectures to detect the hate speech in twitter and specifically racism and sexism.”
MethodText Classification Improved by Integrating Bidirectional LSTM with Two-dimensional Max Pooling
Zhou, P., Qi, Z., Zheng, S., Xu, J., Bao, H., & Xu, B. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1611.06639 · Cited by 434 · cited 1×
“• The neural architecture :We adopted the same architecture of CNN,LSTM,BLSTM cited in [17] , [18], [19]”
Method
Research works printed but not linked: 1
- girshick, 2017, Computer Vision and Deep Learning
Cited by
A deep neural network based multi-task learning approach to hate speech detection
Kapil, P., & Ekbal, A. (2020) · Knowledge-Based Systems · doi:10.1016/j.knosys.2020.106458 · Cited by 184
Pamungkas, E. W., & Patti, V. (2019) · Annual Meeting of the Association for Computational Linguistics: Student Research Workshop (ACL SRW) · doi:10.18653/v1/p19-2051 · Cited by 106
“First, we exploit LSTM with 4 Marwa et al. (2018) claimed to get a higher result, but that paper did not give a complete information about system configuration they used.”
MethodA Review on Deep-Learning-Based Cyberbullying Detection
Hasan, M. T., Hossain, M. A. E., Mukta, M. S. H., Akter, A., Ahmed, M., & Islam, S. (2023) · Future Internet · doi:10.3390/fi15050179 · Cited by 103
“…double LSTM cells are used, takes longer to train, and easy to overfit Convolutional Neural Networks (CNN) [43] Social media content [5–7, Twitter [2,14,25,63,67,136,137], Formspring.me [25,137], Facebook [136], Chats [14], YouTube and Wikipedia [25] Image processing, and object detection…”
BackgroundMethodInfluentialA Human-Centered Systematic Literature Review of Cyberbullying Detection Algorithms
Kim, S., Razi, A., Stringhini, G., Wisniewski, P. J., & De Choudhury, M. (2021) · Proceedings of the ACM on Human-Computer Interaction · doi:10.1145/3476066 · Cited by 79
“…than theoretically-driven – since these features can be easily extracted from online data (e.g., using the LIWC program [120]: [27, 28, 142]) or because of the availability of a variety of off-the-shelf language modeling tools (e.g., word embeddings: [17, 28, 93, 139, 167, 168, 168, 169, 171]).”
BackgroundInfluentialAudio-based Toxic Language Classification using Self-attentive Convolutional Neural Network
Yousefi, M., & Emmanouilidou, D. (2021) · European Signal Processing Conference (EUSIPCO) · doi:10.23919/eusipco54536.2021.9616001 · Cited by 96
“Cyberbullying [3], cyber-harassment [4], abuse [5], hate speech [6], and toxic language [7] are examples of common negative online behavior on different social networking platforms.”
Background
Metadata
- Authors
- Marwa, T., Salima, O., & Souham, M.
- Affiliations
- Université Constantine 2 (DZ)
- Venue
- 2018 3rd International Conference on Pattern Analysis and Intelligent Systems (PAIS)
- Publisher
- IEEE
- Published
- 2018-10
- Pages
- 1-5
- OpenAlex
- W2907447547
- Semantic Scholar
- f67d1c00657e102dd540fd96e2914245a40003c6
- Cited by
- 35
- References
- 22
- Citations by year
- 2026: 1 · 2025: 3 · 2024: 4 · 2023: 6 · 2022: 5 · 2021: 9 · 2020: 5 · 2019: 2
- FWCI
- 1.63
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- International Conference on Pattern Analysis and Intelligent Systems (PAIS) (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Hate Speech and Cyberbullying Detection
- Physical Sciences › Computer Science › Information Systems › Spam and Phishing Detection
- Physical Sciences › Computer Science › Artificial Intelligence › Sentiment Analysis and Opinion Mining
- Keywords
- Twitter, tweets, deep learning, text classification, LSTM, bidirectional LSTM, convolutional neural networks, social media abuse
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Deep learning for online harassment detection in tweets. https://app.cochat.ai/papers/10.1109/pais.2018.8598530
- The paper
- Marwa, T., Salima, O., & Souham, M. (2018). Deep learning for online harassment detection in tweets. 2018 3rd International Conference on Pattern Analysis and Intelligent Systems (PAIS). https://doi.org/10.1109/pais.2018.8598530
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.