Classification of Abusive Comments in Social Media using Deep Learning

Anand, Mukul; Eswari, R.

doi:10.1109/iccmc.2019.8819734

Cited by 40 publications

(6 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We as a whole need a more modest model since English language has around 13 million word in the word reference this is very immense for a model. (CBOW) calculation is chipping away at character level data Mukul Anand, Dr.R.Eswan [9], In this paper the author utilizes Kaggle's poisonous remark dataset for preparing the profound learning model and the information is classified in unsafe, dangerous, gross, hostile, stigmatize and manhandle. On dataset different profound learning strategies get performed and that assists with investigating which profoundlearning procedures is better.…”

Section: Literature Surveymentioning

confidence: 99%

A Review on Online Public Shaming on Social Media: Detection, Analysis and Mitigation

Mane,

Khule,

Nanaware

et al. 2023

Int. J. Res. Publ. Rev.

View full text Add to dashboard Cite

Lately, public shaming on internet based informal organizations and related web-based public gatherings, for example, Twitter has expanded. These events are known to appallingly affect the casualty's social, political, and monetary prosperity. Notwithstanding the undeniable unfortunate results, little has been finished to address this in well known web-based virtual entertainment, with the support that the enormous volume and variety of such comments requires an infeasible number of human arbitrators to finish the work. We mechanize the assignment of identifying public shaming by means of Twitter according to the point of view of casualties in this examination, zeroing in on two perspectives: occasions and shamers. Oppressive, correlation, condemning, strict/ethnic, mockery/joke, and whataboutery are the six kinds of disgraceful tweets, and each tweet is arranged into one of these classifications or as non-shaming. It has been shown that most of clients who post remarks in a shaming occasion are probably going to embarrass the person in question. Shockingly, shamers' Twitter adherent counts develop speedier than those of non-shamers. At last, utilizing AI methods, for example, Backing Vector Machine and Irregular Backwoods, a web application called block Disgrace was fabricated and carried out in light of the order and characterization ofshaming tweets for on-the-fly quieting/obstructing of shamers manhandling a casualty on Twitter.

show abstract

Section: Literature Surveymentioning

confidence: 99%

A Review on Online Public Shaming on Social Media: Detection, Analysis and Mitigation

Mane,

Khule,

Nanaware

et al. 2023

Int. J. Res. Publ. Rev.

View full text Add to dashboard Cite

show abstract

“…Abusive comments were classified in [44] by using deep learning approaches. For this purpose, the toxic comments from the kaggle dataset were labeled as 'Toxic', 'Obscene', 'Insult' and 'Severe toxic' comments.…”

Section: Literature Reviewmentioning

confidence: 99%

Criminal Behavior Identification Using Social Media Forensics

et al. 2022

View full text Add to dashboard Cite

Human needs consist of five levels, which are: physiological needs, safety needs, love needs, esteem needs and self-actualization. All these needs lead to human behavior. If the environment of a person is positive, healthy behavior is developed. However, if the environment of the person is not healthy, it can be reflected in his/her behavior. Machines are intelligent enough to mimic human intelligence by using machine learning and artificial intelligence techniques. In the modern era, people tend to post their everyday life events on social media in the form of comments, pictures, videos, etc. Therefore, social media is a significant way of knowing certain behaviors of people such as abusive, aggressive, frustrated and offensive behaviors. Behavior detection by crawling the social media profile of a person is a crucial and important idea. The challenge of behavior detection can be sorted out by applying social media forensics on social media profiles, which involves NLP and deep learning techniques. This paper is based on the study of state of the art work on behavior detection, and based on the research, a model is proposed for behavior detection. The proposed model outperformed with an F1 score of 87% in the unigram + bigram class, and in the bigram + trigram class, it gave an F1 score of 88% when compared with models applied on state of the art work. This study is a great benefit to cybercrime and cyber-security agencies in shortlisting the profiles containing certain behaviors to prevent crimes in the future.

show abstract

“…Recently, researchers are trying to develop methods and tools to analyze social media sites like Twitter, Facebook, and Snapchat since these mediums has become integral part of our life (Anand and Eswari, 2019). Studies have already been conducted to detect abusive or offensive comments on social media (Sharif et al, 2021a;Aurpa et al, 2022;Sharif et al, 2020).…”

Section: Related Workmentioning

confidence: 99%

COMBATANT@TamilNLP-ACL2022: Fine-grained Categorization of Abusive Comments using Logistic Regression

Hossain¹,

Bishal²,

Hossain³

et al. 2022

Proceedings of the Second Workshop on Speech and Language Technologies for Dravidian Languages

View full text Add to dashboard Cite

With the widespread usage of social media and effortless internet access, millions of posts and comments are generated every minute. Unfortunately, with this substantial rise, the usage of abusive language has increased significantly in these mediums. This proliferation leads to many hazards such as cyber-bullying, vulgarity, online harassment and abuse. Therefore, it becomes a crucial issue to detect and mitigate the usage of abusive language. This work presents our system developed as part of the shared task to detect the abusive language in Tamil. We employed three machine learning (LR, DT, SVM), two deep learning (CNN+BiLSTM, CNN+BiLSTM with Fast-Text) and a transformer-based model (Indic-BERT). The experimental results show that Logistic regression (LR) and CNN+BiLSTM models outperformed the others. Both Logistic Regression (LR) and CNN+BiLSTM with Fast-Text achieved the weighted F 1 -score of 0.39. However, LR obtained a higher recall value (0.44) than CNN+BiLSTM (0.36). This leads us to stand the 2 nd rank in the shared task competition.

show abstract

Classification of Abusive Comments in Social Media using Deep Learning

Cited by 40 publications

References 4 publications

A Review on Online Public Shaming on Social Media: Detection, Analysis and Mitigation

A Review on Online Public Shaming on Social Media: Detection, Analysis and Mitigation

Criminal Behavior Identification Using Social Media Forensics

COMBATANT@TamilNLP-ACL2022: Fine-grained Categorization of Abusive Comments using Logistic Regression

Contact Info

Product

Resources

About