Cyberbullying Detection in Social Networks Using Bi-GRU with Self-Attention Mechanism
Open accessAbstract
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.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
GRU-Att can more accurately interpret and categorize the behavior of cyberbullying on social media platforms.
“This makes it possible for the model to more accurately interpret and categorize the behavior of cyberbullying on social media platforms [71,123,124].”
From A Review on Deep-Learning-Based Cyberbullying Detection (2023) · cites this paper as [123]
GRU-Att can capture long-term dependencies in sequential data faster than the LSTM model while capturing the important part of the input text.
“The model can capture long-term dependencies in sequential data faster than the LSTM model while capturing the important part of the input text by combining Gated Recurrent Units (GRUs) with attention mechanisms.”
From A Review on Deep-Learning-Based Cyberbullying Detection (2023) · cites this paper as [123]
The gated Recurrent Units with Attention (GRU-Att) model has demonstrated exciting results in detecting bullying behavior in social media text.
“The gated Recurrent Units with Attention (GRU-Att) model has also demonstrated exciting results in detecting bullying behavior in a social media text.”
From A Review on Deep-Learning-Based Cyberbullying Detection (2023) · cites this paper as [123]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017) · arXiv · doi:10.48550/arxiv.1706.03762 · Cited by 120,404 · cited 2× · see citing works
“This layer includes a multi-head self-attention mechanism, which is developed by the original attention mechanism [40] that is defined as Equation (5).”
BackgroundHochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318 · cited 1× · see citing works
“Both GRU [46] and long short-term memory (LSTM) [47] are special variants of RNN with logic gates.”
BackgroundGlove: Global Vectors for Word Representation
Pennington, J., Socher, R., & Manning, C. (2014) · Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.3115/v1/d14-1162 · Cited by 34,071 · cited 1× · see citing works
“Instead of initializing the neural work with random embedding, we set the weights for the embedding layer by using pre-trained GloVe word embeddings [45].”
MethodFocal Loss for Dense Object Detection
Lin, T.-Y., Goyal, P., Girshick, R., He, K., & Dollar, P. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.324 · Cited by 30,581 · cited 1× · see citing works
“For model compiling, we use binary_accuracy as metrics, use Adam with initial learning rate 1×10−3 as optimizer and use focal loss [60] as the loss function.”
MethodLearning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation
Cho, K., van Merrienboer, B., Gulcehre, C., Bahdanau, D., Bougares, F., Schwenk, H., & Bengio, Y. (2014) · Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.3115/v1/d14-1179 · Cited by 25,545 · cited 2× · see citing works
“Both GRU [46] and long short-term memory (LSTM) [47] are special variants of RNN with logic gates.”
BackgroundMethodConvolutional 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 3× · see citing works
“TF-IDF + SVM 0.952 0.945 0.947 Char N-grams + LR [21] 0.947 0.943 0.944 Multi-features + LR [24] 0.955 0.948 0.949 KimCNN [59] 0.946 0.946 0.946 Bi-LSTM + Attention [29] 0.935 0.937 0.936 Proposed Method 0.961 0.960 0.960 Wikipedia [58] TF-IDF + SVM 0.923 0.903 0.910 Char N-grams + LR [21] 0.928 0.904 0.912 Multi-feat…”
MethodInfluentialGreff, K., Srivastava, R. K., Koutnik, J., Steunebrink, B. R., & Schmidhuber, J. (2017) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2016.2582924 · Cited by 6,955 · cited 1× · see citing works
“From the perspective of the recurrent unit structure, GRU uses fewer connections and fewer parameters in all places of networks, so model training and generalizing are easier and faster, especially on small sequence data [50].”
BackgroundCyberbullying: Its nature and impact in secondary school pupils
Smith, P. K., Mahdavi, J., Carvalho, M., Fisher, S., Russell, S., & Tippett, N. (2008) · Journal of Child Psychology and Psychiatry · doi:10.1111/j.1469-7610.2007.01846.x · Cited by 3,639 · cited 1× · see citing works
“…are mainly based on statistics and investigation, which focus on the definition, statistical methods and the impacts of cyberbullying, these studies enhanced the factuality of cyberbullying and made researchers pay more attention to cyberbullying from the perspective of severity [14,15].”
BackgroundLearning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation
Cho, K., van Merrienboer, B., Gulcehre, C., Bahdanau, D., Bougares, F., Schwenk, H., & Bengio, Y. (2014) · arXiv · doi:10.48550/arxiv.1406.1078 · Cited by 3,317 · cited 2× · see citing works
“Both GRU [46] and long short-term memory (LSTM) [47] are special variants of RNN with logic gates.”
BackgroundMethodWhite Privilege: Unpacking the Invisible Knapsack (1989) 1
McIntosh, P. (2019) · Routledge eBooks · doi:10.4324/9781351133791-4 · Cited by 2,993 · see citing works
Automated Hate Speech Detection and the Problem of Offensive Language
Davidson, T., Warmsley, D., Macy, M., & Weber, I. (2017) · Proceedings of the International AAAI Conference on Web and Social Media · doi:10.1609/icwsm.v11i1.14955 · Cited by 2,610 · see citing works
Bullying, Cyberbullying, and Suicide
Hinduja, S., & Patchin, J. W. (2010) · Archives of Suicide Research · doi:10.1080/13811118.2010.494133 · Cited by 2,078 · cited 1× · see citing works
“Researchers also studied an association between cyberbullying victimization and suicidal ideation risk [8–10].”
BackgroundHateful Symbols or Hateful People? Predictive Features for Hate Speech Detection on Twitter
Talat, Z., & Hovy, D. (2016) · North American Chapter of the Association for Computational Linguistics · doi:10.18653/v1/n16-2013 · Cited by 1,919 · cited 7× · see citing works
“Data Availability Statement: The data presented in this study are openly available in reference number [21,24,58].”
BackgroundMethodInfluentialBullies Move Beyond the Schoolyard: A Preliminary Look at Cyberbullying
Patchin, J. W., & Hinduja, S. (2006) · Youth Violence and Juvenile Justice · doi:10.1177/1541204006286288 · Cited by 1,701 · see citing works
An Empirical Exploration of Recurrent Network Architectures
Józefowicz, R., Zaremba, W., & Sutskever, I. (2015) · Cited by 1,459 · see citing works
Deep Learning for Hate Speech Detection in Tweets
Badjatiya, P., Gupta, S., Gupta, M., & Varma, V. (2017) · International World Wide Web Conference (WWW) · doi:10.1145/3041021.3054223 · Cited by 1,253 · cited 2× · see citing works
“Badjatiya et al. [28] performed extensive experiments with multiple deep learning methods including convolutional neural network (CNN), long short-term memory (LSTM) and FastText, combined with gradient boosted decision trees, the performance of deep learning model did slightly improved.”
BackgroundFocal Loss for Dense Object Detection
Lin, T.-Y., Goyal, P., Girshick, R., He, K., & Dollár, P. (2017) · arXiv · doi:10.48550/arxiv.1708.02002 · Cited by 1,227 · see citing works
Comparative Study of CNN and RNN for Natural Language ProcessingPreprint
Yin, W., Kann, K., Yu, M., & Schütze, H. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1702.01923 · Cited by 903 · cited 1× · see citing works
“In most cases, GRU and LSTM have similar performance in model prediction capabilities [48,49].”
BackgroundDetecting Hate Speech on Twitter Using a Convolution-GRU Based Deep Neural Network
Zhang, Z., Robinson, D., & Tepper, J. A. (2018) · Lecture notes in computer science · doi:10.1007/978-3-319-93417-4_48 · Cited by 653 · see citing works
Ex Machina: Personal Attacks Seen at Scale
Wulczyn, E., Thain, N., & Dixon, L. (2017) · Proceedings of the 26th International Conference on World Wide Web · doi:10.1145/3038912.3052591 · Cited by 636 · cited 10× · see citing works
“Data Availability Statement: The data presented in this study are openly available in reference number [21,24,58].”
BackgroundInfluentialDive into Deep LearningPreprint
Zhang, A., Lipton, Z. C., Mu, L., & Smola, A. J. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2106.11342 · Cited by 633 · cited 1× · see citing works
“The output of the GRU cell is the hidden state H t , which can be calculated by the following equations [46,51]: The existence of logic gates has a crucial effect on GRU, two logical gates act like two switches in GRU.”
MethodSelf-harm, suicidal behaviours, and cyberbullying in children and young people: systematic review
John, A., Glendenning, A. C., Marchant, A., Montgomery, P., Stewart, A., Wood, S., Lloyd, K., & Hawton, K. (2018) · Journal of Medical Internet Research · doi:10.2196/jmir.9044 · Cited by 580 · cited 1× · see citing works
“Especially for young people, cyberbullying even could lead them to do self-harm and suicide [7].”
BackgroundCyberbullying Via Social Media
Whittaker, E., & Kowalski, R. M. (2015) · Journal of School Violence · doi:10.1080/15388220.2014.949377 · Cited by 498 · cited 1× · see citing works
“Second, as places where cyberbullying occurs most frequently [11], social network platforms usually have the nature of expressing public and anonymous expression.”
BackgroundCall Attention to Rumors: Deep Attention Based Recurrent Neural Networks for Early Rumor Detection
Chen, T., Li, X., Yin, H., & Zhang, J. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-04503-6_4 · Cited by 402 · see citing works
Chatzakou, D., Kourtellis, N., Blackburn, J., Cristofaro, E. D., Stringhini, G., & Vakali, A. I. (2017) · ACM Web Science Conference (WebSci) · doi:10.1145/3091478.3091487 · Cited by 393 · cited 1× · see citing works
“Chatzakou et al. [25] did similar work.”
Deep Learning for Detecting Cyberbullying Across Multiple Social Media Platforms
Agrawal, S., & Awekar, A. (2018) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-76941-7_11 · Cited by 387 · cited 6× · see citing works
“Agrawal et al. [29] also verified the feasibility of neural networks for detecting cyberbullying.”
BackgroundMethodInfluentialAl-Garadi, M. A., Varathan, K. D., & Ravana, S. D. (2016) · Computers in Human Behavior · doi:10.1016/j.chb.2016.05.051 · Cited by 354 · cited 1× · see citing works
“Traditional machine learning algorithms are widely used in detecting negative forms of human behavior [17], the most common classifier adopted by researchers is the support vector machine (SVM), and the most common feature extraction method is the Bag-of-words (BoW) model [6].”
MethodLearning from Bullying Traces in Social Media
Xu, J., Jun, K.-S., Zhu, X., & Bellmore, A. D. (2012) · Cited by 336 · see citing works
Automatic cyberbullying detection: A systematic review
Rosa, H., Pereira, N., Ribeiro, R., Ferreira, P. C., Carvalho, J. P., Oliveira, S., Coheur, L., Paulino, P., Veiga Simão, A. M., & Trancoso, I. (2019) · Computers in Human Behavior · doi:10.1016/j.chb.2018.12.021 · Cited by 307 · cited 1× · see citing works
“Third, one major problem with cyberbullying research is the lack of standard data [12].”
BackgroundAutomated Hate Speech Detection and the Problem of Offensive LanguagePreprint
Davidson, T., Warmsley, D., Macy, M. W., & Weber, I. G. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1703.04009 · Cited by 284 · cited 8× · see citing works
“It could tell that the model of Multi-features + LR [24] is reasonable for this dataset.”
BackgroundMethodInfluentialAutomatic detection of cyberbullying on social networks based on bullying features
Zhao, R., Zhou, A., & Mao, K. (2016) · International Conference on Distributed Computing and Networking (ICDCN) · doi:10.1145/2833312.2849567 · Cited by 243 · cited 1× · see citing works
“[20] concatenated BoW features and latent semantic features, then trained with a linear SVM.”
MethodSampasa‐Kanyinga, H., Roumeliotis, P., & Xu, H. (2014) · PLoS ONE · doi:10.1371/journal.pone.0102145 · Cited by 216 · see citing works
How Noisy Social Media Text, How Diffrnt Social Media Sources?
Baldwin, T., Cook, P., Lui, M., MacKinlay, A., & Wang, L. (2013) · Cited by 193 · see citing works
Al-Garadi, M. A., Hussain, M. R., Khan, N., Murtaza, G., Nweke, H. F., Ali, I., Mujtaba, G., Chiroma, H., Khattak, H. A., & Gani, A. (2019) · IEEE Access · doi:10.1109/access.2019.2918354 · Cited by 186 · cited 2× · see citing works
“In the last few years, online communication tends to be more and more userdriven, and social network platforms have become the hardest hit area of cyberbullying due to millions of users worldwide, efficient communication and “24/7/365” services [6].”
BackgroundMethodCyberbullying detection on twitter using Big Five and Dark Triad features
Balakrishnan, V., Khan, S., Fernandez, T., & Arabnia, H. R. (2019) · Personality and Individual Differences · doi:10.1016/j.paid.2019.01.024 · Cited by 165 · cited 1× · see citing works
“Balakrishnan et al. [26] explored the roles of user personalities from a psychological perspective, they calculated the Big Five (extraversion, openness, conscientiousness, agreeableness, neuroticism) and Dark Triad (Machiavellianism, psychopathy, narcissism) scores, along with user-based and…”
Hate Speech Detection is Not as Easy as You May Think
Arango, A., Pérez, J. E. P., & Poblete, B. (2019) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3331184.3331262 · Cited by 158 · cited 1× · see citing works
“However, two noteworthy studies [30,31] discussed the limitations of the above”
BackgroundCyberbullying Detection Based on Semantic-Enhanced Marginalized Denoising Auto-Encoder
Zhao, R., & Mao, K. (2016) · IEEE Transactions on Affective Computing · doi:10.1109/taffc.2016.2531682 · Cited by 133 · see citing works
Cheng, L., Li, J., Silva, Y. N., Hall, D. L., & Liu, H. (2019) · ACM International Conference on Web Search and Data Mining (WSDM) · doi:10.1145/3289600.3291037 · Cited by 124 · see citing works
Ma, C., Yang, C., Yang, F., Zhuang, Y., Zhang, Z., Jia, H., & Xie, X. (2018) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2018.8486454 · Cited by 103 · see citing works
Hierarchical Attention Networks for Cyberbullying Detection on the Instagram Social Network
Cheng, L., Guo, R., Silva, Y., Hall, D., & Liu, H. (2019) · Proceedings of the 2019 SIAM International Conference on Data Mining · doi:10.1137/1.9781611975673.27 · Cited by 95 · cited 1× · see citing works
“[35] proposed a hierarchical attention network, in order to distinguish cyberbullying sessions that are typically composed of multiple insulting comments.”
BackgroundMachine learning and semantic analysis of in-game chat for cyberbullying
Murnion, S. D., Buchanan, W. J., Smales, A., & Russell, G. (2018) · Computers & Security · doi:10.1016/j.cose.2018.02.016 · Cited by 89 · cited 1× · see citing works
“Murnion et al. [23] put their eyes on cyberbullying in social games.”
BackgroundThe Association Between Cyberbullying, School Bullying, and Suicidality Among Adolescents
Zaborskis, A., Ilionsky, G., Tesler, R., & Heinz, A. (2018) · Crisis · doi:10.1027/0227-5910/a000536 · Cited by 89 · cited 1× · see citing works
“Researchers also studied an association between cyberbullying victimization and suicidal ideation risk [8–10].”
BackgroundAn abusive text detection system based on enhanced abusive and non-abusive word lists
Lee, H.-S., Lee, H.-R., Park, J.-U., & Han, Y.-S. (2018) · Decision Support Systems · doi:10.1016/j.dss.2018.06.009 · Cited by 84 · cited 1× · see citing works
“Lee et al. [22] designed an abusive text detection system based on abusive and non-abusive word lists.”
MethodPredicting human behavior: The next frontiers
Subrahmanian, V. S., & Kumar, S. (2017) · Science · doi:10.1126/science.aam7032 · Cited by 81 · cited 1× · see citing works
“In the aspect of computational studies, machine learning and deep learning help researchers understand more about human behaviors [16].”
BackgroundLu, N., Wu, G., Zhang, Z., Zheng, Y., Ren, Y., & Choo, K. R. (2020) · Concurrency and Computation: Practice and Experience · doi:10.1002/cpe.5627 · Cited by 76 · cited 1× · see citing works
“Lu et al. [32] proposed a character-level convolutional neural network with shortcuts, their model turned out that char-level features could be learned to overcome spelling errors and intentional obfuscation in online posts.”
BackgroundCyberbullying Detection in Social Networks Using Deep Learning Based Models; A Reproducibility Study
Dadvar, M., & Eckert, K. (2018) · arXiv · doi:10.48550/arxiv.1812.08046 · Cited by 71 · cited 1× · see citing works
“[38,39] performed a reproducibility study with deep learning based models that are conducted in [29], their reproduced experiment showed that deep learning based models with word embedding outperform the machine learning models, they also thought the effect of oversampling method should be assessed.”
BackgroundCyberbullying Detection in Social Networks Using Deep Learning Based Models; A Reproducibility StudyPreprint
Dadvar, M., & Eckert, K. (2018) · arXiv (Cornell University) · doi:10.13140/rg.2.2.16187.87846 · Cited by 71 · see citing works
Cyberbullying Detection in Social Networks Using Deep Learning Based Models
Dadvar, M., & Eckert, K. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-59065-9_20 · Cited by 68 · see citing works
Emmery, C., Verhoeven, B., Pauw, G. D., Jacobs, G., Hee, C. V., Lefever, E., Desmet, B., Hoste, V., & Daelemans, W. M. P. (2020) · Language Resources and Evaluation · doi:10.1007/s10579-020-09509-1 · Cited by 66 · cited 1× · see citing works
“However, two noteworthy studies [30,31] discussed the limitations of the above”
BackgroundAlbadi, N., Kurdi, M., & Mishra, S. (2019) · Social Network Analysis and Mining · doi:10.1007/s13278-019-0587-5 · Cited by 56 · cited 1× · see citing works
“[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.”
MethodComPath: User Interest Mining in Heterogeneous Signed Social Networks for Internet of People
Dhelim, S., Ning, H., & Aung, N. (2020) · IEEE Internet of Things Journal · doi:10.1109/jiot.2020.3037109 · Cited by 33 · cited 1× · see citing works
“Various heterogeneous graph neural networks are often used to understand the social characters of the users, such as the users’ emotions, personality types, and interests [37].”
BackgroundWang, S., Cui, L., Liu, L., Lü, X., & Li, Q. (2020) · IEEE International Conference on Services Computing · doi:10.1109/scc49832.2020.00015 · Cited by 18 · cited 1× · see citing works
“[43], Attention Recurrent Neural Network (AttRNN) is proposed to predict the personality traits, they experimented on a dataset of 110,728 Facebook users with 1,580,284 digital footprints and showed that their model has a better performance compared with the original Bi-GRU.”
MethodMa, C., Yang, C., Yang, F., et al. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1804.04555 · Cited by 13 · cited 1× · see citing works
“This design of the Bi-GRU layer can extract more latent relationships between words than the single GRU layer [52].”
A Deep Learning Approach to Aspect-Based Sentiment Prediction
Alexandridis, G., Michalakis, K., Aliprantis, J., Polydoras, P., Tsantilas, P., & Caridakis, G. (2020) · IFIP advances in information and communication technology · doi:10.1007/978-3-030-49161-1_33 · Cited by 8 · cited 1× · see citing works
“For example, Korovesis et al. [41,42] presented a novel bidirectional LSTM/CNN network with an attention layer in order to capture sentiment features of texts in social networks.”
BackgroundLeveraging aspect-based sentiment prediction with textual features and document metadata
Korovesis, K., Alexandridis, G., Caridakis, G., Polydoras, P., & Tsantilas, P. (2020) · Hellenic Conference on Artificial Intelligence · doi:10.1145/3411408.3411433 · Cited by 4 · cited 1× · see citing works
“[41,42] presented a novel bidirectional LSTM/CNN network with an attention layer in order to capture sentiment features of texts in social networks.”
Background
Flagged first, then by citations
1 of the 56 references couldn’t be matched to a known paper.
Research works printed but not linked: 5
- Nahar, V.; Li, X.; Pang, C. An effective approach for cyberbullying detection. Commun. Inf. Sci. Manag. Eng. 2013, 3, 238-247.
- Chatzakou, D.; Kourtellis, N.; Blackburn, J.; De Cristofaro, E.; Stringhini, G.; Vakali, A. Mean Birds: Detecting Aggression and Bullying on Twitter. In Proceedings of the 2017 ACM on Web Science Conference, Troy, NY, USA, 25-28 June 2017; ACM: New York, NY, USA, 2017; pp. 13-22.
- Cheng, L.; Li, J.; Silva, Y.N.; Hall, D.L.; Liu, H. Xbully: Cyberbullying detection within a multi-modal context. In Proceedings of the Twelfth ACM International Conference on Web Search and Data Mining, Melbourne, Australia, 11-15 February 2019; pp. 339-347.
- Curran Associates, Inc.: Long Beach, CA, USA, 2017; pp. 5998-6008.
- DeAngelis, T. Unmasking racial micro aggressions. Monit. Psychol. 2009, 40, 42.
Cited by
Sorted by citations
Zarzycki, K., & Ławryńczuk, M. (2021) · Sensors · doi:10.3390/s21165625 · Cited by 137 · see citing works
Murshed, B. A. H., Abawajy, J., Mallappa, S., Saif, M. A. N., & Al-Ariki, H. D. E. (2022) · IEEE Access · doi:10.1109/access.2022.3153675 · Cited by 137 · see citing works
“[20] designed a classification model that combines a selfAttention mechanism and bi-directional Gated Recurrent Unit (Bi-GRU) to detect cyberbullying in tweets.”
A 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 · see citing works
“This makes it possible for the model to more accurately interpret and categorize the behavior of cyberbullying on social media platforms [71,123,124].”
BackgroundTeng, T. H., & Varathan, K. D. (2023) · IEEE Access · doi:10.1109/access.2023.3275130 · Cited by 77 · see citing works
“Using textual features such as the presence of ‘‘bad’’ words (insult, swear, profane word) in making a post to be an act of cyberbullying has its limitation since the explicit existence of these words is not always right to detect cyberbullying [22].”
BackgroundKumar, R., & Bhat, A. (2022) · International Journal of Information Security · doi:10.1007/s10207-022-00600-y · Cited by 45 · see citing works
“These studies [24, 48–50, 53 RNN is facing the "Vanishing Gradient problem"[54].”
BackgroundAhmed, T., Ivan, S., Kabir, M., Mahmud, H., & Hasan, K. (2022) · Social Network Analysis and Mining · doi:10.1007/s13278-022-00934-4 · Cited by 42 · see citing works
“There has been few works in this domain where authors have proposed a model to detect cyberbullying without considering different traits of it from social media texts such as the work of Fang et al. (2021).”
BackgroundResultArtificial Intelligence-Enabled Cyberbullying-Free Online Social Networks in Smart Cities
Al-Marghilani, A. (2022) · International Journal of Computational Intelligence Systems · doi:10.1007/s44196-022-00063-y · Cited by 38 · see citing works
“Fang et al. [14] proposed a comprehensive method integrating the Bi-GRU and the self-attention method.”
MethodFAEO-ECNN: cyberbullying detection in social media platforms using topic modelling and deep learning
Murshed, B. A. H., Suresha, Abawajy, J., Saif, M. A. N., Abdulwahab, H. M., & Ghanem, F. A. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-15372-3 · Cited by 39 · see citing works
“To discover CB in tweets, a classification model using Bi-directional Gated Recurrent Unit (Bi-GRU) was suggested by Fang et al. [27] in order to learn the underlying relationship among (terms) words in combination with the self-attention mechanism to enhance the classification of CB tweets.”
MethodCyberbullying in text content detection: an analytical review
Azumah, S. W., Elsayed, N., ElSayed, Z., & Ozer, M. (2023) · International Journal of Computers and Applications · doi:10.1080/1206212x.2023.2256048 · Cited by 25 · see citing works
Ahmed, H. A., & Alasadi, H. (2024) · Journal of Sensor and Actuator Networks · doi:10.3390/jsan13020022 · Cited by 24 · see citing works
Ensemble Stacking Model for Sentiment Analysis of Emirati and Arabic Dialects
Shamsi, A. A. A., & Abdallah, S. (2023) · Journal of King Saud University - Computer and Information Sciences · doi:10.1016/j.jksuci.2023.101691 · Cited by 24 · see citing works
“The reverse order of the sequence is used by the backward sub-layer, which performs the same function, The elementwise total of the hidden states in the two sub-layers is used to compute a new hidden state at the end of the Bi-GRU layer (Fang et al., 2021).”
InfluentialCyberbullying detection and machine learning: a systematic literature review
Balakrisnan, V., & Kaity, M. (2023) · Artificial Intelligence Review · doi:10.1007/s10462-023-10553-w · Cited by 23 · see citing works
Chinivar, S., Roopa, M. S., Arunalatha, J. S., & Venugopal, K. R. (2022) · Entertainment Computing · doi:10.1016/j.entcom.2022.100544 · Cited by 25 · see citing works
A Review of Deep Learning Models for Detecting Cyberbullying on Social Media Networks
Batani, J., Mbunge, E., Muchemwa, B., Gaobotse, G., Gurajena, C., Fashoto, S. G., Kavu, T. D., & Dandajena, K. (2022) · Lecture notes in networks and systems · doi:10.1007/978-3-031-09073-8_46 · Cited by 22 · see citing works
Detection of Harassment Type of Cyberbullying: A Dictionary of Approach Words and Its Impact
Mahbub, S., Pardede, E., & Kayes, A. S. M. (2021) · Security and Communication Networks · doi:10.1155/2021/5594175 · Cited by 23 · see citing works
“[55] 2021 Twitter, Wikipedia Yes No Neural network with gated recurrent unit None No”
BackgroundDaniel, R., Murthy, T. S., Kumari, C. D. V. P., Lydia, E. L., Ishak, M. K., Hadjouni, M., & Mostafa, S. M. (2023) · IEEE Access · doi:10.1109/access.2023.3326948 · Cited by 21 · see citing works
“Fanget al. [16] concentrated on text-related cyberbullying recognition as it was a typically utilized data carrier in social networks and was a broadly utilized feature in this work.”
BackgroundDeep Learning for Multi-Labeled Cyberbully Detection: Enhancing Online Safety
Islam, N., Haque, R., Pareek, P. K., Islam, M. B., Sajeeb, I. H., & Ratul, M. H. (2023) · International Conference on Data Science and Network Security (ICDSNS) · doi:10.1109/icdsns58469.2023.10245135 · Cited by 19 · see citing works
Natural language processing for social science research: A comprehensive review
Hou, Y., & Huang, J. (2025) · Chinese Journal of Sociology · doi:10.1177/2057150x241306780 · Cited by 30 · see citing works
“Besides, deep learning models are also commonly used for detecting online extremism (Gaikwad et al., 2021), cyberbullying (Fang et al., 2021; Murshed et al., 2022), and fake news (Ajao et al., 2018).”
Singh, N. M., & Sharma, S. K. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-16402-w · Cited by 20 · see citing works
“To sufficiently address the vanishing and exploding gradients problem in DL models, Fang et al. [7] presented a boosted mechanism based on self-attention.”
Deep autoregressive generative models capture the intrinsics embedded in T-cell receptor repertoires
Jiang, Y., & Li, S. C. (2023) · Briefings in Bioinformatics · doi:10.1093/bib/bbad038 · Cited by 16 · see citing works
Shaban, W. M., Dongxi, X., Daef, K. S., & Elbaz, K. (2024) · Atmospheric Pollution Research · doi:10.1016/j.apr.2024.102162 · Cited by 11 · see citing works
Rangel, C. (2022) · Revista Mediterránea de Comunicación · doi:10.14198/medcom.20749 · Cited by 10 · see citing works
Katta, N., Divya, N., Saravanan, B., Kothari, S., Khan, M. A., & Dasari, A. (2024) · International Conference on Data Science and Network Security (ICDSNS) · doi:10.1109/icdsns62112.2024.10691309 · Cited by 8 · see citing works
“The study [10] suggests a model for detecting cyberbullying that combines Bi-GRU and self-attention mechanisms.”
BackgroundSoft Computing Techniques for Detecting Cyberbullying in Social Multimedia Data
Jing, Y., Ma, H., Ansari, A. S., Sucharitha, G., Omarov, B. S., Kumar, S., Mohammadi, M. S., & Alyamani, K. A. Z. (2023) · Journal of Data and Information Quality · doi:10.1145/3604617 · Cited by 6 · see citing works
“The vast majority of the currently available cyberbullying detection systems [12] are reliant on laborious and time-consuming feature extraction techniques.”
BackgroundAzar, A. T., Noori, H. M., Mahlous, A. R., Alkhayyat, A. H., & Ibraheem, I. K. (2024) · Engineering Technology & Applied Science Research · doi:10.48084/etasr.8314 · Cited by 6 · see citing works
Cyberbullying: A BERT Bi-LSTM Solution for Hate Speech Detection
Nouas, S., Boumahdi, F., Madani, A., & Berrhail, F. (2023) · Lecture notes in networks and systems · doi:10.1007/978-3-031-40598-3_7 · Cited by 5 · see citing works
Emotion Recognition: A New Tool for Healthcare Using Deep Learning Algorithms
Mridha, K., Sarker, T., Zaman, R., Shukla, M., Ghosh, A., & Shaw, R. N. (2024) · Lecture notes in electrical engineering · doi:10.1007/978-981-99-8289-9_47 · Cited by 5 · see citing works
Detection of Cyberbullying on User Tweets and Wikipedia Text using Machine Learning
Venu, V. S., Hariharan, S., Reddy, S. M., Kotha, V. V. R., Muthyala, S. S. R., & Kansal, S. (2023) · International Conference on Applied Artificial Intelligence and Computing (ICAAIC) · doi:10.1109/icaaic56838.2023.10140252 · Cited by 5 · see citing works
“According to Michigan Technology Law Review [16], Cybersmile Foundation [17] and authors [18], [19] Twitter and Wikipedia forums are the major sources.”
BackgroundNithyashree, V., Hiremath, B. N., Vanishree, L., Duvvuri, A., Madival, D. A., & Vidyashree, G. L. (2022) · IEEE International Conference on Distributed Computing, VLSI, Electrical Circuits and Robotics (DISCOVER) · doi:10.1109/discover55800.2022.9974631 · Cited by 4 · see citing works
Hyperparameter Tuned Deep Learning Enabled Cyberbullying Classification in Social Media
Duhayyim, M. A., Mohamed, H. G., Alotaibi, S. S., Mahgoub, H., Mohamed, A., Motwakel, A., Zamani, A. S., & Eldesouki, M. I. (2022) · Computers, materials & continua/Computers, materials & continua (Print) · doi:10.32604/cmc.2022.031096 · Cited by 4 · see citing works
“Still, more volume of studies becomes necessary for solving the implicit language which makes identifying cyber-bullying in mass communication a difficult task [10].”
BioBERT-based SNP-trait Associations Extraction from Biomedical Literature
Dehghani, M., Bokharaeian, B., & Yazdanparast, Z. (2023) · International Conference on Computer and Knowledge Engineering · doi:10.1109/iccke60553.2023.10326231 · Cited by 4 · see citing works
“The bidirectional GRU is able to determine the underlying relationships between words both in the forward and reverse directions [47].”
BackgroundDetecting Cyberbullying in Twitter: A Multi-Model Approach
Joseph, V. A., Prathap, B. R., & Kumar, K. P. (2024) · International Conference on Data Engineering and Communication Systems (ICDECS) · doi:10.1109/icdecs59733.2023.10502699 · Cited by 4 · see citing works
Leveraging deep learning to combat cyberbullying on social media
Vijayarani, M., Balamurugan, G., Govindan, R., & Sevak, S. (2025) · Industrial Psychiatry Journal · doi:10.4103/ipj.ipj_47_25 · Cited by 3 · see citing works
“[15] Cyberbullying detection approaches There are several cyberbullying detection approaches, such as machine learning (ML), deep learning (DL), self-supervised learning (SSL), the use of emojis, supervised learning using ML algorithms, lexicon-based, rule-based, and mixed-initiative approaches.”
Search and Rescue Optimization with Machine Learning Enabled Cybersecurity Model
Mengash, H. A., Alzahrani, J. S., Eltahir, M. M. E., Al‐Wesabi, F. N., Mohamed, A., Hamza, M. A., & Marzouk, R. (2022) · Computer Systems Science and Engineering · doi:10.32604/csse.2023.030328 · Cited by 3 · see citing works
Sunardi, S., Herman, H., Fitriah, F., & Mawarni, S. S. (2023) · Jurnal Pengabdian Masyarakat Bumi Raflesia · doi:10.36085/jpmbr.v6i3.5672 · Cited by 4 · see citing works
Chen, X., Chen, Y., & Yin, G. (2025) · Decision Support Systems · doi:10.1016/j.dss.2025.114409 · Cited by 3 · see citing works
Multilingual Depression Detection in Online Social Media Across Eight Indian Languages
Rajderkar, V., & Bhat, A. (2024) · International Conference for Innovation in Technology (INOCON) · doi:10.1109/inocon60754.2024.10511869 · Cited by 3 · see citing works
Analysing Machine Learning Techniques for Cyberbullying Detection: A Review Study
Aziz, S., Usman, M., Azam, A., Ahmad, F., Bilal, M., & Cheema, A. A. (2022) · International Conference on Emerging Technologies · doi:10.1109/icet56601.2022.10004667 · Cited by 3 · see citing works
“[2] ✓ [3] ✓ ✓ [4] ✓ ✓ ✓ [5] ✓ ✓ ✓ [6] ✓ ✓ [7] ✓ ✓ [8] ✓ ✓ [9] ✓ [10] ✓ ✓ [13] ✓ ✓ ✓ ✓ [19] ✓ ✓ [20] ✓ ✓ ✓ [25] ✓ ✓ [26] ✓ ✓ ✓ ✓ [27] ✓ ✓ [28] ✓ ✓ [29] ✓ ✓”
BackgroundJaiswal, T., Attri, V. K., Kumar, A., & Jaiswal, R. N. (2025) · Lecture notes in networks and systems · doi:10.1007/978-981-96-6906-6_4 · Cited by 2 · see citing works
Pan, Y., Zhang, T., Fang, X., Zhu, Y., Yang, C., Zhou, X., & Zhang, F. (2025) · Optics Communications · doi:10.1016/j.optcom.2025.132490 · Cited by 2 · see citing works
A Study on Highly Accurate Swearing Detection Model Based on Multimodal Data
Deng, J., Liu, Y., Peng, L., Tang, P., & You-min, L. (2023) · International Conference on Electronic Information Technology and Smart Agriculture (ICEITSA) · doi:10.1145/3641343.3641390 · Cited by 2 · see citing works
“Attention mechanisms The attention mechanism is utilized during the training process to enhance the models of expressive ability, effectively capture important information in the text, and grasp the semantic focus [22].”
MethodUrdu Toxic Comment Classification With PURUTT Corpus Development
Saeed, H. H., Khalil, T., & Kamiran, F. (2025) · IEEE Access · doi:10.1109/access.2025.3535862 · Cited by 2 · see citing works
“Fang et al. introduced a combination of Bi-GRU and self-attention for cyberbullying classification [32].”
Information Credibility Detection on Social Media X Using Bi-GRU with GloVe and Firefly Algorithm
Purwanto, F. S. S., & Setiawan, E. B. (2024) · IEEE International Conference on Communication, Networks and Satellite (COMNETSAT) · doi:10.1109/comnetsat63286.2024.10861982 · Cited by 2 · see citing works
A Comprehensive Review of Machine Learning and Deep Learning Techniques for Cyberbullying Detection
Jazzar, M., & Duridi, T. (2025) · Smart innovation, systems and technologies · doi:10.1007/978-981-96-2182-8_1 · Cited by 3 · see citing works
Spotted Hyena Optimizer with Deep Learning Driven Cybersecurity for Social Networks
Hilal, A. M., Hashim, A. H. A., Mohamed, H. G., Alharbi, L. A., Nour, M. K., Mohamed, A., Almasoud, A. S., & Motwakel, A. (2022) · Computer Systems Science and Engineering · doi:10.32604/csse.2023.031181 · Cited by 2 · see citing works
SCOUT: Surveillance and Cyber harassment Observation of Unseen Threats
Biswal, S. (2024) · International Conference on Artificial Intelligence, Metaverse and Cybersecurity (ICAMAC) · doi:10.1109/icamac62387.2024.10828792 · Cited by 1 · see citing works
Analyzing Adversarial Strategies and Countermeasures for Cyberbullying Detection
Juarez, M., Abdukhamidov, E., Sandoval, M., Nazari, M., Hall, D. L., Thiruvathukal, G. K., Abuhmed, T., Silva, Y. N., & Abuhamad, M. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-07715-8_9 · Cited by 1 · see citing works
Mahdi, M. A., Arshed, M. A., & Mumtaz, S. (2026) · Mathematics · doi:10.3390/math14010170 · Cited by 1 · see citing works
Bullyspoiler: Detect and Block Cyberbullying Tweet and Bully using Deep Learning
Soundararajan, K., Pathmapriya, S., Rajashree, R., & Sujitha, J. M. J. (2022) · Irish Interdisciplinary Journal of Science & Research · doi:10.46759/iijsr.2022.6209 · Cited by 1 · see citing works
“Experiments ar conducted to check 3 machine learning and a pair of deep learning models that are; (1) GBM, (2) LR, (3) NB, (4) LSTM-CNN and (5) BiLSTM.”
MethodFrancis, C., & Al-Hababi, A. Y. S. (2023) · IEEE International Conference on Computing (ICOCO) · doi:10.1109/icoco59262.2023.10397625 · Cited by 1 · see citing works
“By prioritizing key clinical feature regions, AM algorithms can increase resilience [41].”
BackgroundTime-User Heterogeneous Neural Interaction Network For Cyberbullying Detection
Zhou, G., Lu, X. S., Zhang, R., Liu, H., & Huang, B. (2024) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn60899.2024.10650684 · Cited by 1 · see citing works
“Session Enconding Module employs a hierarchical word-level and sentence-level self-attention network to model the textual features of comments and transform them into latent representations [13] [14] [15].”
MethodAgrawal, P., Kumar, A., & Tripathi, A. K. (2024) · International Conference on Electronics and Sustainable Communication Systems (ICESC) · doi:10.1109/icesc60852.2024.10689949 · Cited by 1 · see citing works
“In 2021, [7] proposed hybrid models based on machine learning and natural language processing techniques for cyberbullying detection, while [8] focused on cyberbullying detection using Bi-GRU with a self-attention mechanism.”
BackgroundDeep Learning Techniques for Cyber Bullying Detection in Social Networks
Selvi, V., Rappai, S., Kalpana, D. Y., Sabaresan, V., Hegde, N. P., & Sathya, V. (2025) · International Conference on Automation and Computation (AUTOCOM) · doi:10.1109/autocom64127.2025.10957657 · Cited by 1 · see citing works
“Moreover, these kernel sizes are much bigger than those used in order to run the parallel convolution layer cascade [11] - [13].”
Intelligent Cyberbullying Detection by CNN-BiGRU Using Word2Vec and GloVe Word Embeddings
Fattahi, J., Sghaier, F., Mejri, M., Ghayoula, R., & Bahroun, S. (2024) · IIAI International Congress on Advanced Applied Informatics (IIAI-AAI) · doi:10.1109/iiai-aai-winter65925.2024.00049 · Cited by 1 · see citing works
A New Model for a Secure Social Media Application
Riad, K., & Elhoseny, M. (2022) · Applied Sciences · doi:10.3390/app12136375 · Cited by 1 · see citing works
A Global Social Network Values DatasetPreprint
Horan, T. J. (2023) · doi:10.31235/osf.io/6rqjm
Cyberbullying Severe Classification Using Deep Learning Approach
Mohammed, I., & Prasad, R. (2023) · Lecture notes in networks and systems · doi:10.1007/978-981-99-5088-1_31
Regac: Multi-Class Offensive Content Identification Using Graphical ApproachPreprint
Chinivar, S., Roopa, M. S., Arunalatha, J. S., & Venugopal, K. R. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.4772816
Cyberbullying Image Classification using Artificial Intelligence for Safer Online Platform
Krishna, M. V., Verma, R. A., & Kirubanantham, P. (2024) · International Conference on Pervasive Computing and Social Networking (ICPCSN) · doi:10.1109/icpcsn62568.2024.00079
Hate Speech Detection: Recent Advancements and Emerging Technologies
Gupta, P., Gupta, S., Goel, L., Yadav, V., & Singh, D. P. (2024) · Lecture notes in networks and systems · doi:10.1007/978-981-97-4228-8_46
Identification of social media cyberbullying with the aid of machine learning
Elankavi, R., Geethavani, B., Meghana, S. C., Rekha, V., Priya, J. S., & Lookmaan, K. M. (2024) · International Conference on Communication, Computing and Energy Efficient Technologies (I3CEET) · doi:10.1109/i3ceet61722.2024.10993773
Cyberbullying Detection: A Review of the Literature
Nouas, S., & Boumahdi, F. (2024) · International Conference of the African Federation of Operational Research Societies (AFROS) · doi:10.1109/afros62115.2024.11037168
“In cyberbullying detection, pre-trained word embeddings are commonly used [51] [14] [17].”
InfluentialCyberbullying Detection on Social Media using a Hybrid Machine Learning Model
Priyadharsini, P. A., & Geetha, R. (2025) · International Conference on Research Methodologies in Knowledge Management, Artificial Intelligence and Telecommunication Engineering (RMKMATE) · doi:10.1109/rmkmate64874.2025.11042815
Hua, L., Hamid, A. H. A., & Alias, B. S. (2025) · Journal of Computational Methods in Sciences and Engineering · doi:10.1177/14727978251355794
Gupta, A., Matta, P., & Pant, B. (2025) · International Journal of Information Technology · doi:10.1007/s41870-025-02872-8
Cyberbullying Impact Prediction Using Deep Learning Models
Devi, K. N., Rajasekar, V., Shanthi, S., & Chandru, A. (2025) · Lecture notes in electrical engineering · doi:10.1007/978-981-96-9971-1_17
Hartanto, R. S., Lisnanti, E. F., & Fahrazi, M. (2025) · Jurnal Abdi Masyarakat · doi:10.30737/jaim.v9i1.6963
“Misalnya, Cheng et al. (2021) mengembangkan model hierarchical attention networks untuk mendeteksi pola temporal cyberbullying, sedangkan Fang et al. (2021) menerapkan pendekatan Bi-GRU dengan self-attention mechanism dalam mendeteksi interaksi bermuatan negatif di media sosial.”
Sasongko, E. C., & Saputri, P. D. (2025) · International Conference on Artificial Intelligence and Technological Solutions (ICAITech) · doi:10.1109/icaitech66481.2025.11387495
“The Gated Recurrent Unit (GRU) is designed to address the vanishing gradient problem and is simpler than the Long Short-Term Memory (LSTM) model [17].”
OTFS channel estimation method based on IBO-dynamic gated Bi-GRU
Hou, J., Wei, Z., Ji, Y., & Deng, X. (2026) · Scientific Reports · doi:10.1038/s41598-026-44747-3
Sathea Sree, S., & Nalini Joseph, L. (2026) · International Research Journal of Multidisciplinary Technovation · doi:10.54392/irjmt26315
“[15] Bi-GRU with self-attention mechanism Scored significantly better than baselines across all parameters.”
Nadeem, M., & Hongsong, C. (2026) · Computer Science Review · doi:10.1016/j.cosrev.2026.101009
Interpretable Cyberbullying Identification Using Large Language Models
Yin, C., & Chen, Y. (2026) · International Conference on Computational Intelligence and Machine Learning (CIML) · doi:10.1109/ciml69155.2026.11692790
“The comparative baseline models are Bi-GRU+Self-Attention [6] and BERT-finetuned [7] .”
Nguyen, L. V., Le, D. H., Nguyen, V. H., Thuy-Trang, P. H. A. M., Nguyen, R. V., & Lee, O. (2026) · PeerJ Computer Science · doi:10.7717/peerj-cs.4127
Exploring Deep Learning Methods for Accurate ECG Arrhythmia Classification
Wele, I., Khan, D. A., & Kotambkar, D. M. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-10303-1_27
- Authors
- Fang, Y., Yang, S., Zhao, B., & Huang, C.
- Affiliations
- Sichuan University (CN)
- China Electronics Technology Group Corporation (CN)
- Corresponding author
- Huang, Cheng
- Venue
- Information
- Publisher
- MDPI AG
- Published
- 2021-04-16
- Volume
- 12
- Issue
- 4
- Pages
- 171
- ISSN
- 2078-2489 (ISSN-L)
- OpenAlex
- W3153712291
- Cited by
- 74
- References
- 56
- Citations by year
- 2026: 7 · 2025: 15 · 2024: 17 · 2023: 19 · 2022: 14 · 2021: 2
- FWCI
- 4.60
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Information (publishedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- National Key Research and Development Program of China (2018YFB0804503)
- Sichuan Science and Technology Program (2020YFG0076)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Hate Speech and Cyberbullying Detection
- Physical Sciences › Computer Science › Artificial Intelligence › Sentiment Analysis and Opinion Mining
- Physical Sciences › Computer Science › Information Systems › Spam and Phishing Detection
- Keywords
- cyberbullying detection, social networks, bidirectional gated recurrent unit, self-attention, text classification, class imbalance
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
