Modeling the Detection of Textual Cyberbullying

By Karthik Dinakar, Roi Reichart, Henry Lieberman. The Social Mobile Web, 2011

632 citations recorded by the source corpus.

The scourge of cyberbullying has assumed alarming proportions with an ever-increasing number of adolescents admitting to having dealt with it either as a victim or as a bystander. Anonymity and the lack of meaningful supervision in the electronic medium are two factors that have exacerbated this social menace. Comments or posts involving sensitive topics that are personal to an individual are more likely to be internalized by a victim, often resulting in tragic outcomes. We decompose the overall detection problem into detection of sensitive topics, lending itself into text classification sub-problems. We experiment with a corpus of 4500 YouTube comments, applying a range of binary and multiclass classifiers. We find that binary classifiers for individual labels outperform multiclass classifiers. Our findings show that the detection of textual cyberbullying can be tackled by building individual topic-sensitive classifiers.

Subfields: Artificial Intelligence, Information Systems, Social Psychology

View Paper: https://www.semanticscholar.org/paper/812183caa91bab9c2729de916ee3789b68023f39

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Information Systems

In Social Psychology

Citations in the corpus

Studying Generalisability across Abusive Language Detection Datasets (2019, neutral): “(Nobata et al., 2016; Djuric et al., 2015; Warner and Hirschberg, 2012), YouTube (Dinakar et al., 2011), and Wikipedia (Wulczyn et al., 2017), with annotation typically carried out on crowdsourcing platforms such as CrowdFlower (Figure Eight) and Amazon Mechanical Turk.”
Improved cyberbullying detection using gender information (2012, background): “Cyberbullying is defined as an aggressive, intentional act carried out by a group or individual, using electronic forms of contact (e.g. email and chat rooms), repeatedly or over time, against a victim who cannot easily defend her-self [2].”
Ex Machina: Personal Attacks Seen at Scale (2016, methodology): “e project where they experienced the harassment [23]. Online hate speech and cyberbullying are also closely connected to suppressing the expression of others [21], physical violence [29], and suicide [4]. Automated Detection. There have been a number of recent papers on detecting forms of toxic beha”
Challenges in discriminating profanity from hate speech (2017, background): “For example, in analyzing the errors committed by their bullying classiﬁer, Dinakar, Reichart, and Lieberman (2011) note that “the lack of profanity or negativity [can] mislead the classiﬁer” in some cases.”
Toward Multimodal Cyberbullying Detection (2017, methodology): “In our work, we consider textual features indicative of emotion, gender speciﬁc terminology, sexual connotations, as well as the relative distribution of different parts of speech [40, 36, 11, 41, 13, 15].”
Supervised machine learning for the detection of troll profiles in twitter social network: application to a real case of cyberbullying (2015, background): “[13] applied a range of binary and multiclass classifiers on a manually labelled corpus of YouTube comments.”
Developing an online hate classifier for multiple social media platforms (2020, methodology): “In the context of online hate detection, NB has been applied at least by Dinakar et al. [80], Chen et al. [81], and Kwok and Wang [82].”
Understanding Abuse: A Typology of Abusive Language Detection Subtasks (2017, background): “(Dinakar et al., 2011) G e n e r a li z e d “I am surprised they reported on this crap who cares about another dead n*gger?””
SemEval-2019 Task 6: Identifying and Categorizing Offensive Language in Social Media (OffensEval) (2019, background): “Examples of offensive content investigated in previous studies include hate speech (Davidson et al., 2017; Malmasi and Zampieri, 2017, 2018), cyber-bulling (Dinakar et al., 2011), and aggression (Kumar et al., 2018).”
A Unified Deep Learning Architecture for Abuse Detection (2018, background): “For example, there have been various works attempting to detect hate speech [2, 12, 40, 42], cyberbullying [6, 11, 16], and abusive behavior in general [7, 10, 27].”