Don't follow me: Spam detection in Twitter

By Alex Hai Wang. International Conference on Security and Cryptography, 2010

646 citations recorded by the source corpus.

The rapidly growing social network Twitter has been infiltrated by large amount of spam. In this paper, a spam detection prototype system is proposed to identify suspicious users on Twitter. A directed social graph model is proposed to explore the “follower” and “friend” relationships among Twitter. Based on Twitter's spam policy, novel content-based features and graph-based features are also proposed to facilitate spam detection. A Web crawler is developed relying on API methods provided by Twitter. Around 25K users, 500K tweets, and 49M follower/friend relationships in total are collected from public available data on Twitter. Bayesian classification algorithm is applied to distinguish the suspicious behaviors from normal ones. I analyze the data set and evaluate the performance of the detection system. Classic evaluation metrics are used to compare the performance of various traditional classification methods. Experiment results show that the Bayesian classifier has the best overall performance in term of F-measure. The trained classifier is also applied to the entire data set. The result shows that the spam detection system can achieve 89% precision.

Subfields: Artificial Intelligence, Computer Networks and Communications, Information Systems

View Paper: https://www.semanticscholar.org/paper/b03d9d578ce63574014134a2ad6a6f84a15cb9a6

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Networks and Communications

In Information Systems

Citations in the corpus

SybilFuse: Combining Local Attributes with Global Structure to Perform Robust Sybil Detection (2018, neutral): “These approaches span a large category of schemes, including blacklisting [24], whitelisting [25], URL ﬁltering [7], as well as various machine learning methods, such as Bayesian reasoning, Support Vector Machines, and clustering [26]–[28].”
Detecting Spammers on Twitter (2010, background): “Similarly, Wang [26] collected thousands users on Twitter and used classification to distinguish the suspicious behaviors from normal ones.”
Random Walk Based Fake Account Detection in Online Social Networks (2017, background): “In particular, most studies in this direction [5, 19–22, 24, 27] treat Sybil detection as a supervised learning problem; they extract various features from user-generated contents, behaviors, and local graph structure, and then learn machine learning classifiers using a training dataset consisting o”
Structure-Based Sybil Detection in Social Networks via Local Rule-Based Propagation (2018, neutral): “, ego-network) [24], [25] and side information (e.”
SybilSCAR: Sybil detection in online social networks via local rule based propagation (2017, neutral): “, ego-network) [19], [20], [21] and side information (e.”
Detecting Rumors from Microblogs with Recurrent Neural Networks (2016, result): “Our work is also related to, but different from, studies detecting spammers [Lee et al., 2013; Wang, 2010; Hu et al., 2013] and fake images on Twitter [Gupta et al.”
On the Interpretation of Digital Trace Data in Communication and Social Computing Research (2014, background): “After empiricism, the next most prevalent justification strategy was citation (Suh, Hong, Pirolli, & Chi, 2010; A. H. Wang, 2010; Weng et al., 2010).”
Twitter spammer detection using data stream clustering (2014, methodology): “Among the 3239 users, 208 were spam accounts from [15].”
Cybercrime detection in online communications: The experimental case of cyberbullying detection in the Twitter network (2016, methodology): “Recent studies have achieved immense success in applying the machine learning technique to detect spam messages in social media (Chen, Chandramouli, & Subbalakshmi, 2014; McCord & Chuah, 2011; Wang, 2010a; Zheng, Zeng, Chen, Yu, & Rong, 2015).”
Sentiment in Twitter events (2011, background): “The knowledge that big events typically associate with small increases in negative sentiment and sometimes with positive sentiment should aid Twitter Spam detection (e.g., Wang, 2010).”
Review: machine learning techniques applied to cybersecurity (2019, methodology): “[104] designed a system to classify Twitter users that includes a Bayesian classifier to detect text patterns; decision trees, neural networks, SVM and a Bayesian classifier were also used to identify spam bots on Twitter, showing the latter the best performance [105, 106]; Martinez-Romo and Araujo ”
Correlating financial time series with micro-blogging activity (2012, neutral): “Wang [30] and Benevenuto et al.”