A Framework for Early Detection of Antisocial Behavior on Twitter Using Natural Language Processing

Singh, Ravinder; Du, Jiahua; Zhang, Yanchun; Wang, Hua; Miao, Yuan; Sianaki, Omid Ameri; Ulhaq, Anwaar

doi:10.1007/978-3-030-22354-0_43

Cited by 15 publications

(11 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Though there is some sort of mechanism in place through which victims can report such behaviour to the underlying platform, most of these cases go unnoticed as victims are often reluctant to report due to scare of retaliation by the culprit [55], [56]. Even though some victims may report such behaviour to a platform, to manually curtail antisocial behaviour online is a laborious and inconceivable endeavour; therefore, an automatic system is required that can work at a scale [8]. In an effort to promote free speech, most online social media platforms fail to curb online antisocial behaviour.…”

Section: Antisocial Behavior and Social Mediamentioning

confidence: 99%

See 1 more Smart Citation

Deep Learning for Multi-Class Antisocial Behavior Identification From Twitter

Singh

Subramani

et al. 2020

IEEE Access

Self Cite

View full text Add to dashboard Cite

Social Media has become an integral part of our daily life. Not only it enables collaboration and flow of information but has also become an imperative tool for businesses and governments around the world. All this makes a compelling case for everyone to be on some sort of online social media platform. However, this virtuousness is overshadowed by some of its shortcomings. The manifestation of antisocial behaviour online is a growing concern that hinders participation and cultivates numerous social problems. Antisocial behaviour exists in its various forms such as aggression, disregard for safety, lack of remorse, unlawful behaviour, etc. The paper introduces a deep learning-based approach to detect and classify online antisocial behaviour (ASB). The automatic content classification addresses the issue of scalability, which is imperative when dealing with online platforms. A benchmark dataset was created with multi-class annotation under the supervision of a domain expert. Extensive experiments were conducted with multiple deep learning algorithms and their superior results were validated against the results from the traditional machine learning algorithms. Visually enhanced interpretation of the classification process is presented for model and error analyses. Accuracy of up to 99% in class identification was achieved on the ground truth dataset for empirical validation. The study is an evidence of how the cutting-edge deep learning technology can be utilized to solve a real-world problem of curtailing antisocial behaviour, which is a public health threat and a social problem.

show abstract

Section: Antisocial Behavior and Social Mediamentioning

confidence: 99%

“…In our prior work [8], we proposed an approach for binary classification of online antisocial behaviour. Based on the content, the posts were classified as either antisocial or general/non-antisocial.…”

Section: Introductionmentioning

confidence: 99%

Deep Learning for Multi-Class Antisocial Behavior Identification From Twitter

Singh

Subramani

et al. 2020

IEEE Access

Self Cite

View full text Add to dashboard Cite

show abstract

“…Online social media has transformed information consumers into information producers. This phenomenon has enticed researchers from various disciplines to study online social media as an important source of data to explore human behaviour in the physical world [26][27][28][29]. Social media generates an unprecedented amount of data and has led to new ways to discover urban functions and human behaviour [11].…”

Section: Social Media Network Datamentioning

confidence: 99%

Investigation of Social Behaviour Patterns using Location-Based Data – A Melbourne Case Study

Singh¹,

Zhang²,

Wang³

et al. 2018

ICST Transactions on Scalable Information Systems

Self Cite

View full text Add to dashboard Cite

Location-based social networks such as Swarm provide a rich source of information on human behaviour and urban functions. Our analysis of data created by users who voluntarily used check-ins with a mobile application can give insight into a user's mobility and behaviour patterns. In this study, we used location-sharing data from Swarm to explore spatiotemporal, geo-temporal and behaviour patterns within the city of Melbourne. Moreover, we used several tools for different datasets. We used the MeaningCloud tool for sentiment analysis and the LIWC15 tool for psychometric analysis. Also, we employed SPSS software for the descriptive statistical analysis on check-in data to reveal meaningful trends and attain a deeper understanding of human behaviour patterns in the city. The results show that most people do not express strong negative or positive emotions in relation to the places they visit. Behaviour patterns vary based on gender. Furthermore, mobility patterns are different on different days of the week as well as at different times of a day but are not necessarily influenced by the weather.

show abstract

“…For example, Gaffney analyzes more than 3000 tweets by using hashtag #IranElections using graphical methods and user networks and frequencies of top keyword categories to quantify online activities from Iran elections at that time [18]. Similar studies have been conducted by social scientists on the field of healthcare, athletic, antisocial behavior, child abuse and hay fever detection [19,20,21,22,23,24,25].…”

Section: Literature Reviewmentioning

confidence: 99%

Social Networks Mining and Analysis of Specific Groups (Political and Regional) by using API

Khan

Saqib²,

Soomro³

et al. 2018

ICST Transactions on Scalable Information Systems

View full text Add to dashboard Cite

People have informal conversations on social media sites (Twitter, Facebook, etc) that shed light on the current issues around the world-opinions, and concerns about the learning process. Data from such environments can provide valuable information to predict the future to make effective decision making, however, the usage of such data is challenging. The complexity of such data from social media content requires the use of human understanding as well as social network analysis to make predictions. We focused on political groups around the world to discover linkages behind the Terrorism around the World to make predictions depend upon the data available. In this research paper, we focused on tweets related to the politicians and communities based on hashtags like #Syria #Weapons #Terrorist, etc. to identify the linkages of the user who are talking about these issues depending upon the HashTag we used for the content analysis of social network by using Gephi to make a prediction.

show abstract

A Framework for Early Detection of Antisocial Behavior on Twitter Using Natural Language Processing

Cited by 15 publications

References 15 publications

Deep Learning for Multi-Class Antisocial Behavior Identification From Twitter

Deep Learning for Multi-Class Antisocial Behavior Identification From Twitter

Investigation of Social Behaviour Patterns using Location-Based Data – A Melbourne Case Study

Social Networks Mining and Analysis of Specific Groups (Political and Regional) by using API

Contact Info

Product

Resources

About