Detect Me If You Can: Spam Bot Detection Using Inductive Representation Learning

Alhosseini, Seyed Ali; Tareaf, Raad Bin; Najafi, Pejman; Meinel, Christoph

doi:10.1145/3308560.3316504

Cited by 69 publications

(46 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Stanton et al [25] utilized generative adversarial network for spam bot detection. Alhosseini et al [1] proposed a model based on graph convolutional networks for spam bot detection to leverage both node features and neighborhood information.…”

Section: Social Media Bot Detectionmentioning

confidence: 99%

“…Another important aim of our proposed TwiBot-20 is to provide a stable benchmark that evaluates bot detectors' ability to identify diversified bots that co-exist on online social media. To prove that achieving good performance on one type of bot doesn't necessarily indicate the ability to identify diversified bots, we train the community-based bot detector Alhosseini et al [1] on only one of the four interest domains in TwiBot-20 and evaluate it on the full test set. We present its performance in Figure 7.…”

Section: User Diversity Studymentioning

confidence: 99%

“…Figure 7 illustrates that when the method is trained on one of the user interest domains, Alhosseini et al [1] fails to match its performance when it is trained on the full TwiBot-20. As a result, TwiBot-20 could better evaluate bot detection measures in that it contains diversified bots and genuine users, which demands bot detectors to jointly capture different types of bots rather than being limited to a specific bot detection scenario.…”

Section: User Diversity Studymentioning

confidence: 99%

See 2 more Smart Citations

TwiBot-20: A Comprehensive Twitter Bot Detection Benchmark

Feng,

Wan,

Wang

et al. 2021

Preprint

View full text Add to dashboard Cite

Twitter has become one of the thriving social media platforms since its launch in 2006, while complaints about Twitter bots are constantly on the rise. Since then, many efforts have been devoted to the task of Twitter bot detection and the creation of relevant datasets. However, existing benchmarks generally suffer from low levels of user diversity, limited user information and data scarcity. Therefore, these datasets are not sufficient to train and stably benchmark bot detection measures. To alleviate these problems, we present TwiBot-20, a massive Twitter bot detection benchmark, which contains 229,573 users, 33,488,192 tweets, 8,723,736 user property items and 455,958 follow relationships. TwiBot-20 covers diversified bots and genuine users to better represent the real-world Twittersphere. TwiBot-20 also includes three modals of user information to support both binary classification of single users and community-aware approaches. To the best of our knowledge, TwiBot-20 is the largest Twitter bot detection benchmark to date. We reproduce competitive bot detection methods and conduct a thorough evaluation on TwiBot-20 and two other public datasets. Experiment results demonstrate that existing bot detection measures fail to match their previously claimed performance on TwiBot-20, which suggests that Twitter bot detection remains a challenging task and requires further research efforts.

show abstract

Section: Social Media Bot Detectionmentioning

confidence: 99%

Section: User Diversity Studymentioning

confidence: 99%

Section: User Diversity Studymentioning

confidence: 99%

See 1 more Smart Citation

TwiBot-20: A Comprehensive Twitter Bot Detection Benchmark

Feng,

Wan,

Wang

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…At present, a graph convolution neural network has been applied in many fields, such as in a recommendation system [23][24][25][26], in malicious account detection [27], etc. Ali et al [28] proposed a spam bot detection model based on a graph convolutional neural network (GCNN) by using the features of nodes and aggregating the features of node neighborhoods. In addition to the feature set, they also considered the social graph, which can detect spam bots better.…”

Section: Graph Convolutional Networkmentioning

confidence: 99%

“…For conventional methods, the MLP and BP methods are compared using the experimental results in the paper of Ali [28]. For the RF method, we use the default parameter settings from the original literature.…”

Section: Parameter Settingsmentioning

confidence: 99%

An Attention-Based Graph Neural Network for Spam Bot Detection in Social Networks

Zhao

Yang

et al. 2020

Applied Sciences

View full text Add to dashboard Cite

With the rapid development of social networks, spam bots and other anomaly accounts’ malicious behavior has become a critical information security problem threatening the social network platform. In order to reduce this threat, the existing research mainly uses feature-based detection or propagation-based detection, and it applies machine learning or graph mining algorithms to identify anomaly accounts in social networks. However, with the development of technology, spam bots are becoming more advanced, and identifying bots is still an open challenge. This paper proposes a new semi-supervised graph embedding model based on a graph attention network for spam bot detection in social networks. This approach constructs a detection model by aggregating features and neighbor relationships, and learns a complex method to integrate the different neighborhood relationships between nodes to operate the directed social graph. The new model can identify spam bots by capturing user features and two different relationships among users in social networks. We compare our method with other methods on real-world social network datasets, and the experimental results show that our proposed model achieves a significant and consistent improvement.

show abstract