%0 Conference Paper %1 mccord2011detection %A McCord, M. %A Chuah, M. %B Proceedings of the 8th International Conference on Autonomic and Trusted Computing %C Berlin, Heidelberg %D 2011 %E Calero, Jose M. Alcaraz %E Yang, Laurence Tianruo %E Mármol, Félix Gómez %E García-Villalba, Luis Javier %E Li, Xiaolin Andy %E Wang, Yan %I Springer-Verlag %K %P 175--186 %T Spam Detection on Twitter Using Traditional Classifiers %U http://wbox0.cse.lehigh.edu/~chuah/publications/atc11_spam_camera.pdf %X Social networking sites have become very popular in recent years. Users use them to find new friends, updates their existing friends with their latest thoughts and activities. Among these sites, Twitter is the fastest growing site. Its popularity also attracts many spammers to infiltrate legitimate users' accounts with a large amount of spam messages. In this paper, we discuss some user-based and content-based features that are different between spammers and legitimate users. Then, we use these features to facilitate spam detection. Using the API methods provided by Twitter, we crawled active Twitter users, their followers/ following information and their most recent 100 tweets. Then, we evaluated our detection scheme based on the suggested user and content-based features. Our results show that among the four classifiers we evaluated, the Random Forest classifier produces the best results. Our spam detector can achieve 95.7% precision and 95.7% F-measure using the Random Forest classifier. %@ 978-3-642-23495-8

@inproceedings{mccord2011detection, abstract = {Social networking sites have become very popular in recent years. Users use them to find new friends, updates their existing friends with their latest thoughts and activities. Among these sites, Twitter is the fastest growing site. Its popularity also attracts many spammers to infiltrate legitimate users' accounts with a large amount of spam messages. In this paper, we discuss some user-based and content-based features that are different between spammers and legitimate users. Then, we use these features to facilitate spam detection. Using the API methods provided by Twitter, we crawled active Twitter users, their followers/ following information and their most recent 100 tweets. Then, we evaluated our detection scheme based on the suggested user and content-based features. Our results show that among the four classifiers we evaluated, the Random Forest classifier produces the best results. Our spam detector can achieve 95.7% precision and 95.7% F-measure using the Random Forest classifier.}, acmid = {2035717}, added-at = {2016-11-25T11:30:52.000+0100}, address = {Berlin, Heidelberg}, author = {McCord, M. and Chuah, M.}, biburl = {https://www.bibsonomy.org/bibtex/2b764eb3a69c9473bceeafb62117c3b64/nosebrain}, booktitle = {Proceedings of the 8th International Conference on Autonomic and Trusted Computing}, editor = {Calero, Jose M. Alcaraz and Yang, Laurence Tianruo and Mármol, Félix Gómez and García-Villalba, Luis Javier and Li, Xiaolin Andy and Wang, Yan}, interhash = {c4f1d9ce945e00601824c85bc23ffacb}, intrahash = {b764eb3a69c9473bceeafb62117c3b64}, isbn = {978-3-642-23495-8}, keywords = {}, location = {Banff, Canada}, numpages = {12}, pages = {175--186}, publisher = {Springer-Verlag}, series = {Lecture Notes in Computer Science}, timestamp = {2016-11-25T11:30:52.000+0100}, title = {Spam Detection on Twitter Using Traditional Classifiers}, url = {http://wbox0.cse.lehigh.edu/~chuah/publications/atc11_spam_camera.pdf}, year = 2011 }