A Performance Evaluation of Machine Learning-Based Streaming Spam Tweets Detection

Chen, Chao; Zhang, Jun; Xie, Yi; Xiang, Yang; Zhou, Wanlei; Hassan, Mohammad Mehedi; Alelaiwi, Abdulhameed; Alrubaian, Majed

doi:10.1109/tcss.2016.2516039

Cited by 109 publications

(72 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In our group's previous work [2,21], it is observed that Twitter spams are drifting over time in the statistical feature space. The problem is named "twitter spam drift", which seriously affects the detection performance of existing machine learning-based methods.…”

Section: Related Workmentioning

confidence: 93%

Statistical Detection of Online Drifting Twitter Spam

Liu

Zhang

Xiang

2016

Proceedings of the 11th ACM on Asia Conference on Computer and Communications Security

Self Cite

View full text Add to dashboard Cite

Spam has become a critical problem in online social networks. This paper focuses on Twitter spam detection. Recent research works focus on applying machine learning techniques for Twitter spam detection, which make use of the statistical features of tweets. We observe existing machine learning based detection methods suffer from the problem of Twitter spam drift, i.e., the statistical properties of spam tweets vary over time. To avoid this problem, an effective solution is to train one twitter spam classifier every day. However, it faces a challenge of the small number of imbalanced training data because labelling spam samples is time-consuming. This paper proposes a new method to address this challenge. The new method employs two new techniques, fuzzy-based redistribution and asymmetric sampling. We develop a fuzzy-based information decomposition technique to re-distribute the spam class and generate more spam samples. Moreover, an asymmetric sampling technique is proposed to re-balance the sizes of spam samples and non-spam samples in the training data. Finally, we apply the ensemble technique to combine the spam classifiers over two different training sets. A number of experiments are performed on a real-world 10-day ground-truth dataset to evaluate the new method. Experiments results show that the new method can significantly improve the detection performance for drifting Twitter spam.

show abstract

Section: Related Workmentioning

confidence: 93%

Statistical Detection of Online Drifting Twitter Spam

Liu

Zhang

Xiang

2016

Proceedings of the 11th ACM on Asia Conference on Computer and Communications Security

Self Cite

View full text Add to dashboard Cite

show abstract

“…3 to report the account by selecting the reason. Another way which is commonly reported in the literature is mentioning spammers to the official "@spam" account [28,29,37,[56][57][58] but according to the recent report by Twitter, this method of reporting spam is outdated [30]. Also, Wang reports that this method is abused by both hoaxes and spam [29].…”

Section: B How Twitter Deals With Spammentioning

confidence: 99%

A Survey of Spam Detection Methods on Twitter

Kabakuş¹,

Kara²

2017

ijacsa

View full text Add to dashboard Cite

“…Although such a wearable cyber physical system is helpful to analyze user's health condition, i.e., whether a user is already infected or not, it lacks sufficient social information to infer the spread of infectious diseases, i.e., whether the user has a high probability to get infected from others. Social network can offer various applications to mine users' social data during their social interactions [7], [8]. For example, the built-in face-tagging function of Facebook application can identify user's face in pictures and infer if certain users have close social relationships; Wechat friend discovery program can find users in the physical proximity and record social interactions; speech recognition can help to detect if some people cough or sneeze.…”

Section: Introductionmentioning

confidence: 99%

Exploiting Social Network to Enhance Human-to-Human Infection Analysis without Privacy Leakage

Liang

Shen

2018

IEEE Trans. Dependable and Secure Comput.

View full text Add to dashboard Cite

Human-to-human infection, as a type of fatal public health threats, can rapidly spread in a human population, resulting in a large amount of labor and health cost for treatment, control and prevention. To slow down the spread of infection, social network is envisioned to provide detailed contact statistics to isolate susceptive people who has frequent contacts with infected patients. In this paper, we propose a novel human-to-human infection analysis approach by exploiting social network data and health data that are collected by social network and e-healthcare technologies. We enable the social cloud server and health cloud server to exchange social contact information of infected patients and user's health condition in a privacy-preserving way. Specifically, we propose a privacy-preserving data query method based on conditional oblivious transfer to guarantee that only the authorized entities can query users' social data and the social cloud server cannot infer anything during the query. In addition, we propose a privacy-preserving classification-based infection analysis method that can be performed by untrusted cloud servers without accessing the users' health data. The performance evaluation shows that the proposed approach achieves higher infection analysis accuracy with the acceptable computational overhead. Index Terms-Social network, infection analysis, privacy preservation! 1545-5971 (c)

show abstract

A Performance Evaluation of Machine Learning-Based Streaming Spam Tweets Detection

Cited by 109 publications

References 23 publications

Statistical Detection of Online Drifting Twitter Spam

Statistical Detection of Online Drifting Twitter Spam

A Survey of Spam Detection Methods on Twitter

Exploiting Social Network to Enhance Human-to-Human Infection Analysis without Privacy Leakage

Contact Info

Product

Resources

About