Social media bot detection with deep learning methods: a systematic review

Abstract

Social bots are automated social media accounts governed by software and controlled by humans at the backend. Some bots have good purposes, such as automatically posting information about news and even to provide help during emergencies. Nevertheless, bots have also been used for malicious purposes, such as for posting fake news or rumour spreading or manipulating political campaigns. There are existing mechanisms that allow for detection and removal of malicious bots automatically. However, the bot landscape changes as the bot creators use more sophisticated methods to avoid being detected. Therefore, new mechanisms for discerning between legitimate and bot accounts are much needed. Over the past few years, a few review studies contributed to the social media bot detection research by presenting a comprehensive survey on various detection methods including cutting-edge solutions like machine learning (ML)/deep learning (DL) techniques. This paper, to the best of our knowledge, is the first one to only highlight the DL techniques and compare the motivation/effectiveness of these techniques among themselves and over other methods, especially the traditional ML ones. We present here a refined taxonomy of the features used in DL studies and details about the associated pre-processing strategies required to make suitable training data for a DL model. We summarize the gaps addressed by the review papers that mentioned about DL/ML studies to provide future directions in this field. Overall, DL techniques turn out to be computation and time efficient techniques for social bot detection with better or compatible performance as traditional ML techniques.

Gender in the disclosure of loneliness on Twitter during COVID-19 lockdowns

“…Next, we attempted to remove spam or disingenuous users by considering their posting rate ( 46 ). As each tweet we gather contains information about the user at the time of the posting, we use this data to calculate their posting rate: number of tweets they have posted divided by the number of days the account has been active (using user “created at” and “user statuses count” fields).…”

Quantum-Mechanical Modelling of Asymmetric Opinion Polarisation in Social Networks

“…Bot-detection is the process of identifying and distinguishing between automated bots and human users [125,126]. Bots can be detected using algorithms based on behavioural analysis, challenge-response authentication systems, and machine learning techniques [125].…”

“…Bots can be detected using algorithms based on behavioural analysis, challenge-response authentication systems, and machine learning techniques [125]. Nevertheless, despite the constant progress in the domain of cybersecurity, bots remain a serious threat, since they can shape public opinion by spreading fake news and serving as an instrument for cyberbullying and harassment [125,126].…”

“…Comparing such models with the decision-making patterns of users of social networks, a bot-detection system may differentiate a real user from a bot. This functionality should be especially useful, since the developers of bots employ artificial intelligence systems that learn to accurately reproduce certain behaviour patterns of social network users [126]. However, even the most sophisticated bot system cannot yet reproduce human thoughts, thereby revealing the synthetic nature of the bot when its behaviour is benchmarked against a quantum digital twin of a human.…”

Analyzing the worldwide perception of the Russia-Ukraine conflict through Twitter

“…In particular, as stated by Hayawi et al CNNs typically excel in classifying textual content and identifying abusive language, angry terms, and named entities. On the other hand, LSTMs and RNNs are particularly valuable for understanding the relationship between consecutive time point data [15]. Among these, it is worth mentioning the approach outlined in [16] involves the integration of three LSTM models and a fully connected layer.…”