Semantic Text Analysis for Detection of Compromised Accounts on Social Networks

Seyler, Dominic; Li, Lunan; Zhai, ChengXiang

doi:10.1109/asonam49781.2020.9381432

Cited by 5 publications

(3 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…VanDam, Tang, and Tan (2017) performs a measurement study and derives content-based features. Seyler, Li, and Zhai (2020) proposes a method that divides the tweet space randomly into compromised/benign tweets and uses the difference in language distributions as features. Karimi et al (2018) utilizes LSTM networks to capture temporal dependencies to detect compromised accounts.…”

Section: Related Workmentioning

confidence: 99%

“…To counteract this, it is common in supervised learning frameworks to balance datasets to learn a better discriminative function. Balanced datasets were previously used for deleted and suspended account detection (Volkova and Bell 2017), compromised account detection (Karimi et al 2018;VanDam et al 2019;Seyler, Li, and Zhai 2020) and spam detection (Benevenuto et al 2010;Lee, Caverlee, and Webb 2010;Nilizadeh et al 2017;Adewole et al 2019). To create a balanced dataset (i.e.…”

Section: Datasetmentioning

confidence: 99%

See 1 more Smart Citation

Textual Analysis and Timely Detection of Suspended Social Media Accounts

Seyler

Tan

et al. 2021

ICWSM

View full text Add to dashboard Cite

Suspended accounts are high-risk accounts that violate the rules of a social network. These accounts contain spam, offensive and explicit language, among others, and are incredibly variable in terms of textual content. In this work, we perform a detailed linguistic and statistical analysis into the textual information of suspended accounts and show how insights from our study significantly improve a deep-learning-based detection framework. Moreover, we investigate the utility of advanced topic modeling for the automatic creation of word lists that can discriminate suspended from regular accounts. Since early detection of these high-risk accounts is crucial, we evaluate multiple state-of-the-art classification models along the temporal dimension by measuring the minimum amount of textual signal needed to perform reliable predictions. Further, we show that the best performing models are able to detect suspended accounts earlier than the social media platform.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Datasetmentioning

confidence: 99%

Textual Analysis and Timely Detection of Suspended Social Media Accounts

Seyler

Tan

et al. 2021

ICWSM

View full text Add to dashboard Cite

show abstract

“…Existing efforts have the following three limitations. First, they mainly follow the classic big data processing pipeline [32,32,67,85,87]: all social media streams are first collected by distributed log collection systems, transporting massive quantities of users' microblog stream data from geographically distributed web servers to a centralized storage tier. Only after all stream data has arrived do the data processing engines (e.g., Spark Streaming with MLlib [11]) start processing these data.…”

Section: Introductionmentioning

confidence: 99%

Achieving Online and Scalable Information Integrity by Harnessing Social Spam Correlations

Liu

Guan

et al. 2023

IEEE Access

View full text Add to dashboard Cite

Malicious web links, social rumors, fraudulent advertisements, faked comments, and biased propaganda are overwhelmingly influencing online social networks. Enabling information integrity is a hot topic in both academia and industry. Traditional social spam detection techniques rely on centralized processing, focusing only on one specific set of data sources, thereby ignoring the social spam correlations between distributed data sources. In this paper, we propose an online and scalable misinformation detection system, named Spiral, to uncover social spam by leveraging the correlations between different social data sources in geo-distributed sites. The key insight in our approach is to amplify the effectiveness of state-ofthe-art techniques to detect inappropriate posts by enabling the efficient large-scale propagation of detection information across domains. The novelty of our design lies in three key components: (1) a decentralized distributed hash-table-based tree overlay deployment for harvesting and uncovering deceptive information spreading in multiple online social networks communities; (2) a progressive aggregation tree for collecting the properties of these posts and creating new classifiers to actively filter out the propagation of inappropriate posts; and (3) a group communication structure that allows multiple groups to exchange the correlations among distributed social data sources. We designed and implemented a prototype of the Spiral system. Our large-scale experiments, using real-world social data, demonstrate Spiral's scalability, effective loadbalancing, and efficiency in online spam detection for social networks.

show abstract