Discovery and Classification of Twitter Bots

Shevtsov, Alexander; Oikonomidou, Maria; Antonakaki, Despoina; Pratikakis, Polyvios; Kanterakis, Alexandros; Ioannidis, Sotiris; Fragopoulou, Paraskevi

doi:10.1007/s42979-022-01154-5

Cited by 6 publications

(2 citation statements)

References 71 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Twitter has taken steps to identify and remove bot accounts, but it is likely that some bot accounts still exist on the platform. Some studies have estimated that a significant portion of Twitter accounts are bots, but the exact percentage is uncertain 69 – 71 ; thus, it cannot be ruled out that some tweets in the present sample were produced by bots. An estimation made nearly ten years ago indicated that about 8.5% of user accounts were bots.…”

Section: Discussionmentioning

confidence: 81%

Digital inclusion: A mixed-method study of user behavior and content on Twitter

Salzmann-Erikson

2023

DIGITAL HEALTH

View full text Add to dashboard Cite

Objective This study is the first to explore user behavior and characterize the content shared about digital inclusion on Twitter. Methods This mixed-methods research consists of 14,000 tweets featuring the hashtag “#digitalinclusion,” posted on Twitter over 15 months. A machine learning technique, latent Dirichlet allocation, was utilized to discover abstract topics within the tweets statistically. The algorithm identified important keywords and text associated with each topic by modeling the underlying word co-occurrence patterns in the dataset. A manual qualitative content analysis was applied to the qualitative data (1000 tweets). Results Tweets containing #digitalinclusion are driven by four motives: 1) warning against the risks of digital exclusion; 2) tweets that promote actions to increase digital inclusion; 3) tweets that call for others to take action to improve digitalization; and 4) tweets that are neutral but fuel the debate by being active. Quantitative analysis revealed that users discussing digital inclusion come from various continents, including the USA, Europe, Africa, and Asia. There were 3931 unique user accounts, with individuals posting between one and 368 tweets. Approximately half of the tweets contained some embedded media. Conclusion The study concludes that digital inclusion is a subject that engages Twitter users worldwide. Tweets that were associated with community and local initiatives and sustainable development had the highest engagement in terms of the number of retweets and likes. The interpretation is that digital inclusion is crucial for achieving equity in living conditions and enhancing access to health information and services. While initiatives to increase digital inclusion are underway, Twitter users call for more efforts to prevent growing digital exclusion. Twitter, as a social media platform, is valuable for studying the motivations that drive digital inclusion and help counter digital exclusion.

show abstract

Section: Discussionmentioning

confidence: 81%

Digital inclusion: A mixed-method study of user behavior and content on Twitter

Salzmann-Erikson

2023

DIGITAL HEALTH

View full text Add to dashboard Cite

show abstract

“…Con el objetivo de identificar bots dentro de la red de Twitter, se investigó y desarrolló un detector automatizado basado en el comportamiento de las cuentas [Shevtsov et al, 2022;Arin and Kutlu, 2023] . El clasificador propuesto funciona de la siguiente manera:…”

Section: Detección De Bots Sociales Basada En El Comportamiento Del U...unclassified

Botnet activity spotting with artificial intelligence: efficient bot malware detection and social bot identification

Velasco Mata

View full text Add to dashboard Cite

In the cybercrime scope, botnets are networks of bots, automated entities that follow instructions from a cybercriminal. The capacity of these networks to operate en masse has made them one of the most popular tools to carry out malicious activities, from spam distribution to distributed denial-of-service (DDoS) attacks. This has made botnets one of the online threats with the most significant presence, causing billionaire losses to the global economy. The motivation of this PhD Thesis is to research and propose bot detection techniques. Specifically, it is focused on two types of bots: malware bots, i.e., virus programs that can be installed in the victims' devices without their notice; and social bots, i.e., fake accounts in Social Networks that try to masquerade as real humans to deceive regular users.de número de muestras. La propuesta de detector de bots sociales aprovecha BERTopic, un predictor de temas basado en BERT, para clasificar los tuits de los usuarios en 102 categorías. La información resultante se divide en ventanas de tiempo de 15 minutos para caracterizar la actividad de los usuarios y se utiliza para predecir si son bots o humanos mediante el clasificador propuesto: un ensamblado de siete redes neuronales basadas en LSTM. Este sistema obtuvo una precisión de 0,755 y un valor F1 de 0,777 % en el nuevo conjunto de datos.

show abstract

Discovery and Classification of Twitter Bots

et al. 2022

Self Cite

View full text Add to dashboard Cite

Twitter as one of the most popular social networks, offers a means for communication and online discourse, which unfortunately has been the target of bots and fake accounts, leading to the manipulation and spreading of false information. Towards this end, we gather a challenging, multilingual dataset of social discourse on Twitter, originating from 9M users regarding the recent Russo-Ukrainian war, in order to detect the bot accounts and the conversation involving them. We collect the ground truth for our dataset through the Twitter API suspended accounts collection, containing approximately 343K of bot accounts and 8M of normal users. Additionally, we use a dataset provided by Botometer-V3 with 1,777 Varol, 483 German accounts, and 1,321 US accounts. Besides the publicly available datasets, we also manage to collect 2 independent datasets around popular discussion topics of the 2022 energy crisis and the 2022 conspiracy discussions. Both of the datasets were labeled according to the Twitter suspension mechanism. We build a novel ML model for bot detection using the state-of-the-art XGBoost model. We combine the model with a high volume of labeled tweets according to the Twitter suspension mechanism ground truth. This requires a limited set of profile features allowing labeling of the dataset in different time periods from the collection, as it is independent of the Twitter API. In comparison with Botometer our methodology achieves an average 11% higher ROC-AUC score over two real-case scenario datasets.

show abstract

Discovery and Classification of Twitter Bots

Cited by 6 publications

References 71 publications

Digital inclusion: A mixed-method study of user behavior and content on Twitter

Digital inclusion: A mixed-method study of user behavior and content on Twitter

Botnet activity spotting with artificial intelligence: efficient bot malware detection and social bot identification

Discovery and Classification of Twitter Bots

Contact Info

Product

Resources

About