Detecting Linkedin Spammers and its Spam Nets

Spam is one of the main problems of the WWW. Many studies exist about characterising and detecting several types of Spam (mainly Web Spam, Email Spam, Forum/Blob Spam and Social Networking Spam). Nevertheless, to the best of our knowledge, there are no studies about the detection of Spam in Linkedin. In this article, we propose a method for detecting Spammers and Spam nets in the Linkedin social network. As there are no public or private Linkedin datasets in the state of the art, we have manually built a dataset of real Linkedin users, classifying them as Spammers or legitimate users. The proposed method for detecting Linkedin Spammers consists of a set of new heuristics and their combinations using a kNN classifier. Moreover, we proposed a method for detecting Spam nets (fake companies) in Linkedin, based on the idea that the profiles of these companies share content similarities. We have found that the proposed methods were very effective. We achieved an F-Measure of 0.971 and an AUC close to 1 in the detection of Spammer profiles, and in the detection of Spam nets, we have obtained an F-Measure of 1.

@article{Prieto2013,
  title     = {Detecting Linkedin Spammers and its Spam Nets},
  journal   = {International Journal of Advanced Computer Science and Applications},
  volume    = {4},
  number    = {9},
  year      = {2013},
  publisher = {The Science and Information Organization},
  author    = {V´ictor M. Prieto and Manuel A´ lvarez and Fidel Cacheda},
  doi       = {10.14569/IJACSA.2013.040930},
  url       = {https://doi.org/10.14569/IJACSA.2013.040930}
}