The DOI matches its Crossref record; 57 of the 71 references it cites were found in a registry or an index.
Online Human-Bot Interactions: Detection, Estimation, and Characterization
Open accessAbstract
Increasing evidence suggests that a growing amount of social media content is generated by autonomous entities known as social bots. In this work we present a framework to detect such entities on Twitter. We leverage more than a thousand features extracted from public data and meta-data about users: friends, tweet content and sentiment, network patterns, and activity time series. We benchmark the classification framework by using a publicly available dataset of Twitter bots. This training data is enriched by a manually annotated collection of active Twitter users that include both humans and bots of varying sophistication. Our models yield high accuracy and agreement with each other and can detect bots of different nature. Our estimates suggest that between 9% and 15% of active Twitter accounts are bots. Characterizing ties among accounts, we observe that simple bots tend to interact with bots that exhibit more human-like behaviors. Analysis of content flows reveals retweet and mention strategies adopted by bots to interact with different target groups. Using clustering analysis, we characterize several subclasses of accounts, including spammers, self promoters, and accounts that post content from connected applications.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“A decade of research on social bot detection 13-15 has established that automated accounts exhibit distinctive temporal, linguistic, and network signatures 14,31 .”
From The Moltbook Illusion: Separating Human Influence from Emergent Behavior in AI Agent Societies (2026) · cites this paper as [14]
In the citing paper
“Various characteristics, ranging from account metadata to social network structure, are then considered during the process (Ferrara et al., 2016;Varol et al., 2017).”
From Anatomy of an AI-powered malicious social botnet (2024) · cites this paper as Varol et al., 2017
In the citing paper
“These features are shown to be effective in identifying social bots and are described in detail in the literature [42, 43, 48].”
From Detection of Novel Social Bots by Ensembles of Specialized Classifiers (2020) · cites this paper as [43]
Citations
Cites
Scikit-learn: Machine Learning in Python
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Müller, A., Nothman, J., Louppe, G., et al. (2012) · arXiv · doi:10.48550/arxiv.1201.0490 · Cited by 71,016 · cited 1×
“We benchmarked our system using several off-the-shelf algorithms provided in the scikit-learn library (Pedregosa et al. 2011).”
Method(2008) · Journal of Machine Learning Research · Cited by 33,051
Twitter mood predicts the stock market
Bollen, J., Mao, H., & Zeng, X. (2011) · Journal of Computational Science · doi:10.1016/j.jocs.2010.12.007 · Cited by 5,206 · cited 1×
“Sentiment extracted from social media conversations has been used to forecast offline events including financial market fluctuations (Bollen, Mao, and Zeng 2011), and is known to affect information spreading (Mitchell et al. 2013; Ferrara and Yang 2015).”
Recognizing contextual polarity in phrase-level sentiment analysis
Wilson, T., Wiebe, J., & Hoffmann, P. (2005) · doi:10.3115/1220575.1220619 · Cited by 3,388 · cited 1×
“…extraction techniques to generate various sentiment features, including arousal , valence and dominance scores (Warriner, Kuperman, and Brysbaert 2013), happiness score (Kloumann et al. 2012), polarization and strength (Wilson, Wiebe, and Hoffmann 2005), and emoti-con score (Agarwal et al. 2011).”
Ferrara, E., Varol, O., Davis, C., Menczer, F., & Flammini, A. (2016) · Communications of the ACM · doi:10.1145/2818717 · Cited by 1,973
Self-citationNorms of valence, arousal, and dominance for 13,915 English lemmas
Warriner, A. B., Kuperman, V., & Brysbaert, M. (2013) · Behavior Research Methods · doi:10.3758/s13428-012-0314-x · Cited by 1,959
Sentiment Analysis of Twitter Data
Agarwal, A., Xie, B., Vovsha, I., Rambow, O., & Passonneau, R. J. (2011) · Cited by 1,565
Social bots distort the 2016 U.S. Presidential election online discussion
Bessi, A., & Ferrara, E. (2016) · First Monday · doi:10.5210/fm.v21i11.7090 · Cited by 993 · cited 7×
“Features extracted from user meta-data have been used to classify users and patterns before (Mislove et al. 2011; Ferrara et al. 2016a).”
InfluentialSelf-citationBotOrNot: A System to Evaluate Social Bots
Davis, C. A., Varol, O., Ferrara, E., Flammini, A., & Menczer, F. (2016) · Proceedings of the 25th International Conference Companion on World Wide Web - WWW 16 Companion · doi:10.1145/2872518.2889302 · Cited by 909 · cited 2×
“This system leverages more than one thousand features to evaluate the extent to which a Twitter account exhibits similarity to the known characteristics of social bots (Davis et al. 2016).”
BackgroundMethodSelf-citationIs the Sample Good Enough? Comparing Data from Twitter's Streaming API with Twitter's Firehose
Morstatter, F., Pfeffer, J., Liu, H., & Carley, K. (2021) · Proceedings of the International AAAI Conference on Web and Social Media · doi:10.1609/icwsm.v7i1.14401 · Cited by 863 · cited 9×
“For each of these accounts, we collected their tweets through the Twitter Search API.”
MethodInfluential
Cited by
Sorted by citations
Items 501–600 of 933
"23andMe confirms: I'm super white" -- Analyzing Twitter Discourse On Genetic TestingPreprint
Mittos, A., Blackburn, J., & Cristofaro, E. D. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1801.09946 · Cited by 6
“There is one interesting outlier for 23andMe (@littlebytesnews), who liked more than 1M tweets; this is likely to be a bot, as also confirmed by Botome-ter [35].”
BackgroundMethodInfluentialData Poisoning Attacks to Local Differential Privacy ProtocolsPreprint
Cao, X., Jia, J., & Gong, N. Z. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1911.02046 · Cited by 43
“t is a practical threat model to assume that an attacker can inject fake users. For instance, around 5% of Facebook’s monthly active users are fake [3], and 9% to 15% of active Twitter users are fake [55]. Since an LDP protocol executes the encode and perturb steps locally on each user’s device, the attacker has acces…”
BackgroundBastos, M. T., & Mercea, D. (2018) · SSRN Electronic Journal · doi:10.2139/ssrn.3192412 · Cited by 6
BotCF: Improving the Social Bot Detection Performance By Focusing on the Community Features
Liu, F., Li, Z., Yang, C., Gong, D., Liu, F., Ma, R., & Borş, A. G. (2025) · IEEE Transactions on Network and Service Management · doi:10.1109/tnsm.2025.3600474 · Cited by 6
“One statistic found that bot accounts accounted for 15% of Twitter’s active accounts in 2017 [8].”
Ren, X., & Wei, Q. (2024) · Software · doi:10.3390/software3030018 · Cited by 6
“Common vulnerabilities include re-entrancy, integer overflow, and underflow, unchecked return values, and access control issues [12,13].”
BackgroundArroyo‐Machado, W., Herrera‐Viedma, E., & Torres‐Salinas, D. (2025) · Journal of the Association for Information Science and Technology · doi:10.1002/asi.24998 · Cited by 8
“Other studies, such as Varol et al. (2017), estimate a prevalence between 9% and 15% based on the analysis of user activity and machine learning validation.”
On-chain and Off-chain Blockchain Data Collection
Zheng, P., Zheng, Z., Wu, J., et al. (2021) · Springer eBooks · doi:10.1007/978-981-16-0127-9_2 · Cited by 5
Twitter Bot Identification: An Anomaly Detection Approach
Alkulaib, L., Zhang, L., Sun, Y.-S., & Lu, C.-T. (2022) · 2022 IEEE International Conference on Big Data (Big Data) · doi:10.1109/bigdata55660.2022.10020919 · Cited by 6
“Existing bot detection methods on Twitter focus on hand crafted features that identify bots using profile-related and tweet-related features using traditional classification meth-ods [6]–[10].”
MethodFactors affecting Consumer Brand Sabotage virality: a study of an Indian brand #boycott
Bhatia, R., Gupta, A., Vimalkumar, M., & Sharma, D. (2023) · Information Systems and e-Business Management · doi:10.1007/s10257-023-00628-0 · Cited by 7
“Fourthly, we suggest that future scholars analyse online negative brand-related behaviours to compare and contrast performance-based and value-based CBS. Lastly, CBS might also be influenced by fake news and algorithmic bots that incite ideological battles (Varol et al. 2017).”
Zhang, M., Qi, X., Liu, X., & Zhang, K. (2024) · SAGE Open · doi:10.1177/21582440241248891 · Cited by 5
“The developers of Botometer have pointed out that simple social bots score between 0.8 and 1.0, and complex social bots score between 0.5 and 0.7 (Varol et al., 2017).”
InfluentialRetractedFedKG: A Knowledge Distillation-Based Federated Graph Method for Social Bot Detection
Wang, X., Chen, K., Wang, K., Wang, Z., Zheng, K., & Zhang, J. (2024) · Sensors · doi:10.3390/s24113481 · Cited by 5
“Varol et al. [13] evaluated the classification models using over a thousand features extracted from users’ public data and metadata, including friends, tweets, network patterns, activity time series, and sentiment.”
MethodSocialbots Whitewashing Contested Elections; A Case Study from Honduras
Gallagher, E., Suárez–Serrato, P., & Richards, E. I. V. (2018) · Advances in intelligent systems and computing · doi:10.1007/978-981-13-1165-9_50 · Cited by 5
Graber, R., & Lindemann, T. (2018) · Aktivismus- und Propagandaforschung · doi:10.1007/978-3-658-22118-8_3 · Cited by 5
Detecting Twitter Bots That Share SoundCloud Tracks
Bruns, A., Moon, B., Münch, F. V., et al. (2018) · International Conference on Social Media and Society (SMSociety) · doi:10.1145/3217804.3217923 · Cited by 5
“The presence of bots on Twitter has come under increasing scrutiny in recent years; some studies estimate that bots represent up to 15% of Twitter accounts [1].”
BackgroundDeep Temporal Analysis of Twitter Bots
Rajendran, G., Ram, A., Vijayan, V., & Poornachandran, P. (2020) · Communications in computer and information science · doi:10.1007/978-981-15-4301-2_4 · Cited by 5
Increasing Engagement in a Cyber-Awareness Training Game
Wray, R. E., Massey, L., Medina, J., & Bolton, A. E. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-50439-7_10 · Cited by 5
Towards a pragmatic detection of unreliable accounts on social networks
Guimar�ães, N., Figueira, Á., & Torgo, L. (2021) · Online Social Networks and Media · doi:10.1016/j.osnem.2021.100152 · Cited by 5
Li, J.-H., Paananen, V., Suryanarayana, S. A., Huusko, E., Kuutila, M., Mäntylä, M. V., & Hosio, S. (2023) · Conference on Human Information Interaction and Retrieval · doi:10.1145/3576840.3578308 · Cited by 7
“Varol et al. [42] estimated that between 9% and 15% of active Twitter accounts are bots.”
BackgroundMeasuring the Information-Foraging Behaviors of Social Bots Through Word Usage
Stine, Z. K., Khaund, T., & Agarwal, N. (2018) · IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM) · doi:10.1109/asonam.2018.8508811 · Cited by 5
“The Twitter-focused bot detection platform, BotOrNot, uses a Random Forest classifier to estimate the likelihood that an account is an automated bot [2][3][4].”
BackgroundMethodAnalysis of Account Engagement in Onsetting Twitter Message Cascades
Kessling, P., & Grimme, C. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-39627-5_10 · Cited by 5
Bozarth, L., Panda, A., Budak, C., & Pal, J. (2020) · International Conference on Information and Communication Technologies and Development (ICTD) · doi:10.1145/3392561.3394636 · Cited by 5
“high bot activities on Twitter [76], we also randomly sample a subset of (sample size corresponds to 99% confidence-level and +/-3 confidence interval) Narendra Modi and Rahul Gandhi followers and then use Botometer [24] to determine their bot scores (technical details are in the Appendix).”
MethodShahrezaye, M., Meckel, M., Steinacker, L., & Suter, V. (2021) · Advances in intelligent systems and computing · doi:10.1007/978-3-030-73100-7_73 · Cited by 5
“This number is much lower than estimations on general bot activity on Twitter, which is assumed to be up to 15% [Davis et al., 2016; Varol et al., 2017].”
BackgroundMethods and Challenges in Social Bots Detection: A Systematic Review
de Morais, D. M. G., & Digiampietri, L. A. (2021) · Anais do Simpósio Brasileiro de Sistemas de Informação (SBSI) · doi:10.1145/3466933.3466973 · Cited by 5
“Estas características podem ser agrupadas em seis grupos distintos [58]:”
BackgroundInfluentialDöring, N. M. (2018) · Springer eBooks · doi:10.1007/978-3-658-18017-1_8-1 · Cited by 7
Repenser la sociologie du numérique à l’aune de notre vie commune avec les robots sociaux
Glassey, O. (2018) · Sociologie et sociétés · doi:10.7202/1054274ar · Cited by 6
Twitter bot detection & categorization - a comparative study of machine learning methods
Novotný, J. (2019) · Lund University Publications Student Papers (Lund University) · Cited by 5
Don't Feed the Troll: Detecting Troll Behavior via Inverse Reinforcement LearningPreprint
Luceri, L., Giordano, S., & Ferrara, E. (2020) · arXiv (Cornell University) · Cited by 5
Self-citationDetection and Prevention of Abuse in Online Social Networks
Talukder, S. K. (2019) · doi:10.25148/etd.fidc007657 · Cited by 5
Rieder, A., Chakraborty, S., Goyal, S., & Berndt, D. J. (2024) · Journal of Information Technology · doi:10.1177/02683962241280657 · Cited by 4
“…namely, Celebrities , In fl uencers , Broadcasters , Ampli fi ers , Commentators , and Viewers (Krishnamurthy et al., 2008; Tinati et al., 2012; Varol et al., 2017) along with two synthetic agent classes Simple Bots and Sophisticated Bots (Chu et al., 2012; Salge et al., 2022; Varol et al.,…”
Piao, J., Lu, Z., Gao, C., & Li, Y. (2025) · International World Wide Web Conference (WWW) · doi:10.1145/3696410.3714537 · Cited by 4
“Social bots are automated software programs that mimic human behaviors to interact with users on social media platforms [15, 43, 45, 47].”
Perceptions of nature and its non‐material contributions to people at Mount Kilimanjaro
Degano, M. E., Kwaslema, S. A., Böhning‐Gaese, K., Hemp, A., Lehnen, L., Martín‐López, B., Pearson, J., Mueller, T., & Arbieu, U. (2025) · People and Nature · doi:10.1002/pan3.70079 · Cited by 4
He, B., Jiang, X., Wu, Q., Liu, H., Yang, Y., & Liao, Y. (2025) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3711896.3736862 · Cited by 5
Meng, X., Wang, X., & Peng, T. (2026) · Communication Methods and Measures · doi:10.1080/19312458.2026.2613829 · Cited by 7
“Next, supervised learning methods require labeled data to train classifiers for bot detection (Sayyadiharikandeh et al., 2020; Varol et al., 2017), whereas unsupervised learning uses clustering techniques without labeled training data.”
InfluentialHanok, S., & Shankaraiah, S. (2023) · The International Arab Journal of Information Technology · doi:10.34028/iajit/20/2/15 · Cited by 4
Ellis, C. H., Moore, J. B., Ho, P., Ahmed, W., & Evans, C. E. L. (2024) · Social Network Analysis and Mining · doi:10.1007/s13278-024-01404-9 · Cited by 5
“It is estimated that between 9% (Varol et al. 2017) and 29% (Weng and Lin 2022) of social media activity is attributable to bots.”
Zeng, K., Zhao, L., & Wang, X. (2025) · Sensors · doi:10.3390/s25134179 · Cited by 4
Zeng, Z., Li, T., Sun, J., Sun, S., & Zhang, Y. (2022) · Data Technologies and Applications · doi:10.1108/dta-02-2022-0084 · Cited by 4
Agent-based null models for examining experimental social interaction networks
Fennell, S. C., Gleeson, J. P., Quayle, M., et al. (2023) · Scientific Reports · doi:10.1038/s41598-023-32295-z · Cited by 4
Swinging in the States: Does disinformation on Twitter mirror the US presidential election system?
Pratelli, M., Petrocchi, M., Saracco, F., & Nicola, R. D. (2023) · International World Wide Web Conference (WWW) · doi:10.1145/3543873.3587638 · Cited by 4
“Botometer [49, 51] is based on a supervised machine learning approach employing Random Forest classifers [5].”
MethodInfluentialEstrategias de astroturfing en las publicaciones de Tik Tok: el caso de la Ley Trans España
Ballesteros-Aguayo, L., Ribeiro, J. C., & García-Estévez, N. (2023) · Historia y Comunicación Social · doi:10.5209/hics.92238 · Cited by 4
Wilmet, A., & Lamarche-Perrin, R. (2019) · Springer proceedings in complexity · doi:10.1007/978-3-030-14459-3_12 · Cited by 4
“For instance, Varol et al. [17] detect bots by means of a supervised machine learning technique.”
MethodOn Twitter Bots Behaving Badly: Empirical Study of Code Patterns on GitHub
Millimaggi, A., & Daniel, F. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-19274-7_14 · Cited by 4
“[15] propose a bottom-up approach to the identification of bots with similar online behavior.”
BackgroundLi, K., Guo, B., Zhang, Q., Yuan, J., & Yu, Z. (2019) · International World Wide Web Conference (WWW) · doi:10.1145/3308560.3316704 · Cited by 4
“Existing works [15, 16, 17] pay more attention to individual polluter rather than CCP.”
BackgroundKellner, A., Wressnegger, C., & Rieck, K. (2020) · doi:10.1145/3380786.3391399 · Cited by 4
“[28], who claim that between 9 % and 15 % of active Twitter accounts are likely to be operated automatically.”
BackgroundResultSalamanos, N., Jensen, M. J., Iordanou, C., & Sirivianos, M. (2023) · Lecture notes in computer science · doi:10.1007/978-981-99-5177-2_4 · Cited by 4
“95 AUC classification performance [24].”
MethodInfluentialDetecting Social Spammers in Sina Weibo Using Extreme Deep Factorization Machine
Wu, Y., Fang, Y., Shang, S., Wei, L., Jin, J., & Wang, H. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-62005-9_13 · Cited by 5
“The word counts of different posts of a social spammer are usually similar [20].”
InfluentialVisual analysis of global research trends in social bots based on bibliometrics
Chen, X., Gao, S., & Zhang, X. (2021) · Online Information Review · doi:10.1108/oir-06-2021-0336 · Cited by 4
TG-OUT: temporal outlier patterns detection in Twitter attribute induced graphs
Dimitriadis, I., Poiitis, M., Faloutsos, C., & Vakali, A. I. (2022) · World Wide Web · doi:10.1007/s11280-021-00986-0 · Cited by 4
“Anomaly detection is often based on supervised machine learning techniques applied on labelled datasets, at which users are classified into spammers and non-spammers [50], bots or humans [49], cyberbullies [7], etc.”
BackgroundMethodForschungsgemeinschaft, D., & Gredel, E. (2022) · De Gruyter eBooks · doi:10.1515/9783110721447 · Cited by 5
How Many Features Do We Need to Identify Bots on Twitter?
Tabassum, F., Mubarak, S., Liu, L., & Du, J. T. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-28035-1_22 · Cited by 6
Control and spread of contagion in networks with global effects
Higgins, J., & Sabarwal, T. (2023) · Journal of Public Economic Theory · doi:10.1111/jpet.12643 · Cited by 4
Terrorism and War: Twitter Cascade Analysis
Carchiolo, V., Longheu, A., Malgeri, M., Mangioni, G., & Previti, M. (2018) · Studies in computational intelligence · doi:10.1007/978-3-319-99626-4_27 · Cited by 4
“cyborg users), whose goal is to exploit humans social biases to strengthen fakes propagation [5,6].”
BackgroundPaletz, S. B. F., Auxier, B. E., & Golonka, E. (2019) · SpringerBriefs in complexity · doi:10.1007/978-3-030-16413-3_6 · Cited by 4
Semantically-Aware Statistical Metrics via Weighting Kernels
Cresci, S., Pietro, R. D., & Tesconi, M. (2019) · IEEE International Conference on Data Science and Advanced Analytics (DSAA) · doi:10.1109/dsaa.2019.00019 · Cited by 4
“imented with class imbalances frequently confronted when performing bot detection in the wild (bot % ≥ 10%) [34].”
BackgroundРоботизация социальных сетей и ее политические последствия
МИХАЙЛЕНОК, О. М., & Malysheva, G. A. (2020) · Власть · doi:10.31171/vlast.v28i1.7046 · Cited by 4
“Социальные боты – это «контролируемые программным обеспечением учетные записи, которые алгоритмически производят контент и устанавливают взаимодействия с другими пользователями» [Varol et al. 2017: 1].”
Deng, X., Dai, Z., Sun, M., & Lv, T. (2020) · Communications in computer and information science · doi:10.1007/978-981-15-9129-7_17 · Cited by 6
Pietro, R. D., Raponi, S., Caprolu, M., & Cresci, S. (2020) · Advances in information security · doi:10.1007/978-3-030-60618-3_2 · Cited by 4
Detecting Bot Accounts on Twitter by Measuring Message Predictability.
Przybyła, P. (2019) · CLEF (Working Notes) · Cited by 4
Who Tweets in Italian? Demographic Characteristics of Twitter Users
Righi, A., Gentile, M. M., & Bianco, D. M. (2019) · Springer proceedings in mathematics & statistics · doi:10.1007/978-3-030-21158-5_25 · Cited by 5
A Feature-Driven Approach for Identifying Pathogenic Social Media Accounts
Alvari, H., Beigi, G., Sarkar, S., Ruston, S. W. V., Corman, S. R., Davulcu, H., & Shakarian, P. (2020) · International Conference on Data Intelligence and Security (ICDIS) · doi:10.1109/icdis50059.2020.00010 · Cited by 6
“We chose this method over Botometer [42] as it achieved comparable results with far less number of features ([42] uses over 1,500 features)(see also [24]).”
BackgroundMethodInfluentialMuhle, F. (2020) · Ars digitalis · doi:10.1007/978-3-658-30702-8_3 · Cited by 4
Tomaselli, V., Cantone, G. G., & Mazzeo, V. (2022) · Advances in business strategy and competitive advantage book series · doi:10.4018/978-1-7998-9223-6.ch016 · Cited by 3
Threat or Opportunity? - Examining Social Bots in Social Media Crisis CommunicationPreprint
Brachten, F., Mirbabaie, M., Stieglitz, S., et al. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1810.09159 · Cited by 4
“This estimation is based on more than 1.000 features of an account (e.g. the tweet frequency or friend-follower-ratio) and has been applied in past studies for that purpose (Varol et al., 2017).”
BackgroundMethodInfluentialThe Big Data toolkit for Psychologists: Data Sources and MethodologiesPreprint
Peters, H., Marrero, Z., & Gosling, S. D. (2021) · doi:10.31234/osf.io/keqmh · Cited by 4
A Nudge to Credible Information as a Countermeasure to Misinformation: Evidence from TwitterPreprint
Hwang, E. H., & Lee, S. (2021) · SSRN Electronic Journal · doi:10.2139/ssrn.3928343 · Cited by 4
Exposure to social bots amplifies perceptual biases and regulation propensityPreprint
Yan, H. Y., Yang, K., Shanahan, J. E., et al. (2022) · doi:10.31219/osf.io/ap2qf · Cited by 4
Self-citationA novel approach to identify deepfake text using social media data
Devi, B. T., & Rajasekaran, R. (2025) · Social Network Analysis and Mining · doi:10.1007/s13278-025-01536-6 · Cited by 5
“NLP has made bots increasingly deceitful and difficult to identify (Varol et al. 2017; Abokho-dair et al. 2015).”
A Unified Neuro-Fuzzy Framework to Assess the User Credibility on Twitter
Laila, K., Jayashree, P. R., & Vinuvarsidh, V. (2023) · IETE Journal of Research · doi:10.1080/03772063.2023.2220687 · Cited by 3
“According to the report [5], roughly 15% of spam URLs lead to risky sources.”
BackgroundDesign and analysis of tweet-based election models for the 2021 Mexican legislative election
Vigna-Gómez, A., Murillo, J., Ramirez, M., et al. (2023) · EPJ Data Science · doi:10.1140/epjds/s13688-023-00401-w · Cited by 3
“However, different academic studies suggest that the percentage of bots may be as high as ≈15 [58, 59].”
BackgroundImproving Social Bot Detection Through Aid and Training
Kenny, R., Fischhoff, B., Davis, A. L., & Canfield, C. (2023) · Human Factors The Journal of the Human Factors and Ergonomics Society · doi:10.1177/00187208231210145 · Cited by 4
“Twitter has had notably nefarious social bot activity (Chu et al., 2010; Huang, 2020; Timberg & Dwoskin, 2018; Uyheng & Carley, 2020; Varol et al., 2017).”
BackgroundAn Analysis of Social Bot Activity on X in Modern Japan
Ippa, S., Okubo, T., & Hashimoto, M. (2024) · IEEE Access · doi:10.1109/access.2024.3454536 · Cited by 3
“In the English-speaking world, between 9% and 15% of accounts by population are socialbots[18], and the 2016 U.S. presidential election was credited with enabling national and foreign governments to deploy legions of socialbots to influence the direction of the online conversation[14].”
Jin, R., & Liao, Y. (2025) · Applied Sciences · doi:10.3390/app15116230 · Cited by 3
“Since its release, more training datasets and tailored features have been used to improve its performance [13,14].”
Hijacking algorithmic bias: analyzing the political discourse around ChatGPT on social media
Cohen, M., & Toch, E. (2025) · Information Communication & Society · doi:10.1080/1369118x.2025.2561046 · Cited by 6
“Specifically, we looked to retweet-to-like ratios, which suggest amplification over personal endorsement, common in cam-paigns trying to boost visibility rather than create engagement (Ratkiewicz et al., 2011; Varol et al., 2017).”
Deploying South African social honeypots on Twitter
Cornelissen, L. A., Barnett, R. J., Kepa, M. A. M., et al. (2018) · Annual Conference of the South African Institute of Computer Scientists and Information Technologists (SAICSIT) · doi:10.1145/3278681.3278703 · Cited by 3
“There is some nuance introduced in different type of automated agents, such as political bots or spamming bots [13].”
BackgroundUsing Common Enemy Graphs to Identify Communities of Coordinated Social Media Activity
Overbey, L. A., Ek, B., Pinzhoffer, K., et al. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-21741-9_10 · Cited by 3
Introduction to Stylistic Models and Applications
Savoy, J. (2020) · Springer eBooks · doi:10.1007/978-3-030-53360-1_1 · Cited by 3
Hepp, A. (2021) · Springer eBooks · doi:10.1007/978-3-658-32133-8_25 · Cited by 3
Census of Twitter users: Scraping and describing the national network of South Korea
Guan, L., Liu, X. F., Sun, W.-J., Liang, H., & Zhu, J. J. H. (2022) · PLoS ONE · doi:10.1371/journal.pone.0277549 · Cited by 3
“Previous studies on Twitter bots stated that up to 15% of Twitter accounts are bots, not people [22].”
BackgroundData Curation Strategies to Support Responsible Big Social Research and Big Social Data Reuse
Mannheimer, S. (2022) · International Journal of Digital Curation · doi:10.2218/ijdc.v17i1.823 · Cited by 3
“Data quality can be affected by bots and fake accounts; as many as 15% of active Twitter accounts are bots (Varol et al., 2017).”
BackgroundProperties of Malicious Social Bots
Kolomeets, M., & Chechulin, A. A. (2023) · Proceedings of Telecommunication Universities · doi:10.31854/1813-324x-2023-9-1-94-104 · Cited by 3
Enhancing the Detection of Social bots on Twitter using Ensemble machine Learning Technique
Mohanty, S., Dwivedy, S. P., Acharya, A. A., Mohapatra, S., Sahoo, S. S., Samal, S., & Samal, S. (2022) · International Conference on Advancements in Smart, Secure and Intelligent Computing (ASSIC) · doi:10.1109/assic55218.2022.10088372 · Cited by 5
“Varol, It could extract more than 1000 features relative to the account from Twitter API’s data and produce a so-called bot score: the higher the score, the more likely the account is controlled by the software, according to the algorithm [17].”
MethodGrouping of Twitter users according to contents of their tweets
Naznin, F., & Mahanta, A. K. (2023) · Indonesian Journal of Electrical Engineering and Computer Science · doi:10.11591/ijeecs.v31.i2.pp876-884 · Cited by 3
Behaviour and Bot Analysis on Online Social Networks
Alipour, S. A., Orji, R., & Zincir-Heywood, N. (2023) · International Journal of Technology and Human Interaction · doi:10.4018/ijthi.327789 · Cited by 3
“For examining the effectiveness of the features, models have been trained by each of those among which metadata and content have shown the best results (Varol et al., 2017).”
BackgroundMethodInfluentialAl-azawi, R., & Al‐Mamory, S. O. (2024) · INTELIGENCIA ARTIFICIAL · doi:10.4114/intartif.vol27iss73pp142-158 · Cited by 3
“Furthermore, the study achieved a remarkable accuracy rate of 0.99 through the selection of top-ranking features and dimension reduction.”
MethodInfluentialLi, Y., Sun, X., Yang, R., Sun, X., Chen, S., Wang, S., Bhuiyan, M. Z. A., Zomaya, A. Y., & Xu, J. (2024) · International Journal of Machine Learning and Cybernetics · doi:10.1007/s13042-024-02392-0 · Cited by 4
“Early approaches of social anomaly detection [11–13], simply relied on shallow classifiers based on straight-for-ward extraction of static account information, i.e., extracting the screen name, id, and the number of followers from metadata, assuming that early-stage anomalous users tend to adopt…”
Changing Dynamics of Political Party Communication: The Case of New Media in Bangladesh
Sakib, N. H., Shishir, M. F. J., & Rahman, M. S. (2024) · Bandung · doi:10.1163/21983534-11030001 · Cited by 3
Integrating higher-order relations for enhanced twitter bot detection
Reiche, S., Cohen, S., Simonov, K., & Friedrich, T. (2024) · Social Network Analysis and Mining · doi:10.1007/s13278-024-01372-0 · Cited by 3
“1 In 2017, (Varol et al. 2017) estimated that bots make up to 15% of all Twitter accounts (Varol et al. 2017).”
Tweet Credibility Ranker: A Credibility Features’ Fusion Model
Qureshi, K. A., & Malick, R. A. S. (2025) · Cognitive Computation · doi:10.1007/s12559-025-10413-5 · Cited by 3
“Therefore, this feature is about the identification of such accounts that is done through the API [78].”
Technology-facilitated Societal Consensus
Kampik, T., & Najjar, A. (2019) · ACM Conference on User Modeling, Adaptation and Personalization (UMAP) · doi:10.1145/3314183.3323451 · Cited by 3
“(ii ) technical: the use of social bots [13], whose number is estimated to be in the order of tens of millions on Facebook and Twitter [47], can facilitate the spread of fake news on these social networks and allow fake news and pro-paganda publishers to automatize and scale their fake news…”
BackgroundBargar, A., Pitts, S., Butkevics, J., & McCulloh, I. (2019) · doi:10.23919/cycon.2019.8756832 · Cited by 4
“SNA metrics are coupled with other features like post timing, content analysis and user-specific measures that are then fed into ML models to mine interactions and unique characteristics of the specific campaigns [13, 14, 15, 20].”
BackgroundBot-like Behavior Detection in Online Banking
Wang, Y., & Wang, L. (2019) · doi:10.1145/3335484.3335518 · Cited by 3
“Varol compute the entropy, as one of eight statistics of every individual feature [15].”
MethodBot and Gender Detection using Textual and Stylistic Information.
Giachanou, A., & Ghanem, B. (2019) · CLEF (Working Notes) · Cited by 3
Identification of tweets that mention books
Yada, S., Kageura, K., & Paris, C. L. (2019) · International Journal on Digital Libraries · doi:10.1007/s00799-019-00273-4 · Cited by 3
“Varol et al. [71] estimated that between 9 and 15% of Twitter accounts are bots.”
BackgroundSpatiotemporal Filtering Pipeline for Efficient Social Networks Data Processing Algorithms
Mukhina, K. D., Visheratin, A. A., & Nasonov, D. A. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-50433-5_7 · Cited by 3
“It has been proven by many studies that a RF approach is efficient for bots and spam detection [23,26].”
MethodVitkova, L., Kotenko, I. V., Kolomeets, M., Tushkanova, O. N., & Chechulin, A. A. (2020) · Advances in intelligent systems and computing · doi:10.1007/978-3-030-50097-9_42 · Cited by 3
Savoy, J. (2020) · Springer eBooks · doi:10.1007/978-3-030-53360-1_6 · Cited by 5
Dehghan, E., Moon, B., Keller, T. R., Graham, T., Bruns, A., & Angus, D. (2020) · AoIR Selected Papers of Internet Research · doi:10.5210/spir.v2020i0.11199 · Cited by 3
“We relied on Botometer to detect social bots within this dataset (e.g. Varol et al. 2017).”
MethodDie überschätzte Gefahr? Twitter-Bots im Europawahlkampf 2019
Pfaffenberger, F., & Heinrich, P. (2020) · Springer eBooks · doi:10.1007/978-3-658-31472-9_4 · Cited by 3
Using Machine Learning to Inductively Learn Semantic Rules
Jabardi, M. H., & Hadi, A. S. (2021) · Journal of Physics Conference Series · doi:10.1088/1742-6596/1804/1/012099 · Cited by 3
“Threats to users include loss of privacy, identity theft, installation of malware, and Sybil attacks that impugn a real user’s reputation by falsely attributing opin ions to them, and sexual harassment [7].”
BackgroundDynamics of Fake News Diffusion
Chakraborty, T. (2020) · The information retrieval series · doi:10.1007/978-3-030-62696-9_5 · Cited by 3
Metadata
- Authors
- Varol, O., Ferrara, E., Davis, C., Menczer, F., & Flammini, A.
- Affiliations
- Indiana University (US)
- University of Southern California (US)
- Venue
- Proceedings of the International AAAI Conference on Web and Social Media
- Published
- 2017-05-03
- Volume
- 11
- Issue
- 1
- Pages
- 280-289
- ISSN
- 2162-3449 (ISSN-L)
- 2334-0770
- OpenAlex
- W2595521492
- Semantic Scholar
- 6a025e1e4d7269b20dadc91e14f28a93124fc145
- Cited by
- 981
- References
- 71
- Citations by year
- 2026: 90 · 2025: 81 · 2024: 84 · 2023: 105 · 2022: 96 · 2021: 103 · 2020: 164 · 2019: 126 · 2018: 104 · 2017: 28
- FWCI
- 75.86
- Type
- conference paper
- Language
- en
- Open access
- diamond
- doi.org
- Locations
- Proceedings of the International AAAI Conference on Web and Social Media (publishedVersion)
- Funders
- National Science Foundation (1101743)
- James S. McDonnell Foundation
- Defense Advanced Research Projects Agency (W911NF-12-1-0037)
- Office of Naval Research (N15A-020-0053)
- Topics
- Physical Sciences › Computer Science › Information Systems › Spam and Phishing Detection
- Social Sciences › Sociology and Political Science › Misinformation and Its Impacts
- Physical Sciences › Computer Science › Artificial Intelligence › Hate Speech and Cyberbullying Detection
- Keywords
- social bots, bot detection, Twitter, human-bot interaction, botnets, spammer account, self-promotion, user activity patterns, social media
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
