53 Hits in 3.8 sec

Following you home from school: A critical review and synthesis of research on cyberbullying victimization

Robert S. Tokunaga
2010 Computers in Human Behavior  
Cyberbullying victimization is one such offense that has recently received a fair amount of attention.  ...  About 20-40% of all youths have experienced cyberbullying at least once in their lives. Demographic variables such as age and gender do not appear to predict cyberbullying victimization.  ...  Future research on in cyberbullying should pay greater attention to the qualities of the technology through which the cyberbullying takes place, as potential moderators of cyberbullying relationships.  ... 
doi:10.1016/j.chb.2009.11.014 fatcat:pl45l3a2gvcmld2byzhfuqrlnq

Current Limitations in Cyberbullying Detection: on Evaluation Criteria, Reproducibility, and Data Scarcity [article]

Chris Emmery, Ben Verhoeven, Guy De Pauw, Gilles Jacobs, Cynthia Van Hee, Els Lefever, Bart Desmet, Véronique Hoste, Walter Daelemans
2019 arXiv   pre-print
The detection of online cyberbullying has seen an increase in societal importance, popularity in research, and available open data.  ...  These predominantly yield small datasets that fail to capture the required complex social dynamics and impede direct comparison of progress.  ...  We would like to thank Prodromos Ninas, Kostas Stoitsas, and Alejandra Hernández Rejón for carrying out a range of trial experiments on this task, Bram Willemsen for helpful remarks, and in particularÁkos  ... 
arXiv:1910.11922v1 fatcat:umpv23es5zaqfg32co4aqrrbnu

Current limitations in cyberbullying detection: On evaluation criteria, reproducibility, and data scarcity

Chris Emmery, Ben Verhoeven, Guy De Pauw, Gilles Jacobs, Cynthia Van Hee, Els Lefever, Bart Desmet, Véronique Hoste, Walter Daelemans
2020 Language Resources and Evaluation  
AbstractThe detection of online cyberbullying has seen an increase in societal importance, popularity in research, and available open data.  ...  These predominantly yield small datasets that fail to capture the required complex social dynamics and impede direct comparison of progress.  ...  Finally, we want to express our gratitude to all authors providing cyberbullying data open-source.  ... 
doi:10.1007/s10579-020-09509-1 fatcat:d6rr47n3m5hvvfs5r3jgspdbyu

Offensive Language Detection with BERT-based models, By Customizing Attention Probabilities [article]

Peyman Alavi, Pouria Nikvand, Mehrnoush Shamsfard
2021 arXiv   pre-print
Next, we create two separate (because of the differences in the types of the employed datasets) equations based on Offensive Scores for each language to re-distribute the 'Attention Mask' input for paying  ...  more attention to more offensive phrases.  ...  It aims to make the model just paying attention to very offensive tokens.  ... 
arXiv:2110.05133v1 fatcat:n5bcgxdfr5ewtc4qj2uvrgrb6u

A systematic review of Hate Speech automatic detection using Natural Language Processing [article]

Md Saroar Jahan, Mourad Oussalah
2021 arXiv   pre-print
multiplication of social media platforms, which offer anonymity, easy access and online community formation, and online debate, the issue of hate speech detection and tracking becomes a growing challenge to  ...  Besides, as we wanted to pay special attention to machine learning and deep learning-based methods, several related abbreviations Search sources We used two different databases (ACM Digital Library  ...  Likewise, sentence-4, 'John Doe is not a good person,' does not contain Insult words, but it is considered HS, cyberbullying, and offensive.  ... 
arXiv:2106.00742v1 fatcat:qwxjwgma4zaynemge57cu7xqlm


Nugraha Eka, Putra
2013 unpublished
in order to regulate Cyberbullying.  ...  This paper will try to review and analyse regulation of Cyberbullying in the Indonesian Law, are the regulation could accomodate Cyberbullying as a part of cybercrime, or the regulation need to be revised  ...  Acknowledgements The author would like to thank Dr. H. Supriyadi SH., MH.  ... 

Detecting Transaction-based Tax Evasion Activities on Social Media Platforms Using Multi-modal Deep Neural Networks [article]

Lelin Zhang
2020 arXiv   pre-print
Due to such convenient and anarchic nature, they have also been used rampantly to promote and conduct business activities between unregistered market participants without paying taxes.  ...  To build such a tool, we collected a dataset of 58,660 Instagram posts and manually labelled 2,081 sampled posts with multiple properties related to transaction-based tax evasion activities.  ...  Special thanks to Fujitsu Australia Limited for providing the computational resources for this study.  ... 
arXiv:2007.13525v1 fatcat:pwcrtdrugraxhhgfd3apb4kcgq

Ethical and technical challenges of AI in tackling hate speech

Diogo Cortiz, Arkaitz Zubiaga
2021 The International Review of Information Ethics  
As a case study, we used an AI model developed to detect hate speech on social networks, a concept for which varying definitions are given in the scientific literature and consensus is lacking.  ...  Finally, we argue that AI can assist with the detection of hate speech in social media, provided that the final judgment about the content has to be made through a process with human involvement.  ...  We must pay attention to the filters we use in the collection to ensure a more diverse and representative dataset.  ... 
doi:10.29173/irie416 fatcat:3hbvylxjffd6xh44fq7xfpajs4

A Language Model for Misogyny Detection in Latin American Spanish Driven by Multisource Feature Extraction and Transformers

Edwin Aldana-Bobadilla, Alejandro Molina-Villegas, Yuridia Montelongo-Padilla, Ivan Lopez-Arevalo, Oscar S. Sordia
2021 Applied Sciences  
This research contributes to the development of models for the automatic detection of misogynistic texts in Latin American Spanish and contributes to the design of data augmentation methodologies since  ...  Creating effective mechanisms to detect misogyny online automatically represents significant scientific and technological challenges.  ...  As mentioned, one of the main characteristics of BERT is the transformer structure, where its encoder pay attention to the sentence as a whole and not dividing it into tokens.  ... 
doi:10.3390/app112110467 fatcat:hrsqz2oe5rbhhnonbjxgtz5eia

MOMENTA: A Multimodal Framework for Detecting Harmful Memes and Their Targets [article]

Shraman Pramanick, Shivam Sharma, Dimitar Dimitrov, Md Shad Akhtar, Preslav Nakov, Tanmoy Chakraborty
2021 arXiv   pre-print
Although memes are typically humorous, recent days have witnessed an escalation of harmful memes used for trolling, cyberbullying, and abuse.  ...  To solve these tasks, we propose MOMENTA (MultimOdal framework for detecting harmful MemEs aNd Their tArgets), a novel multimodal deep neural network that uses global and local perspectives to detect harmful  ...  Intervention with human moderation would be required in order to ensure that this does not occur. Intended Use.  ... 
arXiv:2109.05184v2 fatcat:ntmq4pv6kjdhvebjyohuikqppe

Detecting and Classifying Malevolent Dialogue Responses: Taxonomy, Data and Methodology [article]

Yangjun Zhang, Pengjie Ren, Maarten de Rijke
2020 arXiv   pre-print
Conversational interfaces are increasingly popular as a way of connecting people to information.  ...  We make three contributions to advance research on this task. First, we present a Hierarchical Malevolent Dialogue Taxonomy (HMDT).  ...  This research pays attention to a certain category and focuses more on multilingual aspects.  ... 
arXiv:2008.09706v1 fatcat:zipkjnxpqvfdzliccpraiaafpa

SemEval-2020 Task 12: Multilingual Offensive Language Identification in Social Media (OffensEval 2020) [article]

Marcos Zampieri, Preslav Nakov, Sara Rosenthal, Pepa Atanasova, Georgi Karadzhov, Hamdy Mubarak, Leon Derczynski, Zeses Pitenis, Çağrı Çöltekin
2020 arXiv   pre-print
The task involves three subtasks corresponding to the hierarchical taxonomy of the OLID schema (Zampieri et al., 2019a) from OffensEval 2019.  ...  A total of 528 teams signed up to participate in the task, 145 teams submitted systems during the evaluation period, and 70 submitted system description papers.  ...  It is also supported by the Tanbih project at the Qatar Computing Research Institute, HBKU, which aims to limit the effect of "fake news," propaganda and media bias by making users aware of what they are  ... 
arXiv:2006.07235v2 fatcat:gqo4hmya2zcxpkcyzusq4hztlu

L-Boost: Identifying Offensive Texts from Social Media Post in Bengali

M. F. Mridha, Md. Anwar Hussen Wadud, Md. Abdul Hamid, Muhammad Mostafa Monowar, M. Abdullah-Al-Wadud, Atif Alamri
2021 IEEE Access  
pay attention to both precision and recall.  ...  on its tokens to find attention from a split word.  ... 
doi:10.1109/access.2021.3134154 fatcat:jaaavefprne2xlukdtywtxzd6a

Hope Speech detection in under-resourced Kannada language [article]

Adeep Hande, Ruba Priyadharshini, Anbukkarasi Sampath, Kingston Pal Thamburaj, Prabakaran Chandran, Bharathi Raja Chakravarthi
2021 arXiv   pre-print
Henceforth, KanHope aims to instigate research in Kannada while broadly promoting researchers to take a pragmatic approach towards online content that encourages, positive, and supportive.  ...  Consequently, we propose creating an English-Kannada Hope speech dataset, KanHope and comparing several experiments to benchmark the dataset.  ...  Certain ethnic groups or individuals fall victim to manipulating social media to foster destructive or disruptive behaviour, a common scenario in cyberbullying [7, 8].  ... 
arXiv:2108.04616v2 fatcat:mxawtdglxbgmflsmjers6htdja

Hate Speech Detection and Racial Bias Mitigation in Social Media based on BERT model [article]

Marzieh Mozafari, Reza Farahbakhsh, Noel Crespi
2020 arXiv   pre-print
Next, we introduce a bias alleviation mechanism in hate speech detection task to mitigate the effect of bias in training set during the fine-tuning of our pre-trained BERT-based model.  ...  pre-trained BERT-based model with the new re-weighted samples.  ...  In summary, we should consider in future studies paying substantial attention to sexual and gender identities as long as dialect and social identity of the speaker in concert with highly correlated n-grams  ... 
arXiv:2008.06460v2 fatcat:lp4jsbv57reatd223mu6p7cluq
« Previous Showing results 1 — 15 out of 53 results