Classification profiles of actor’s information security in social networking services (in the example of microblog Twitter)
Keywords:
social networking service, actor, information security, machine learning, binary classification, threats, evaluationAbstract
Social networking services (SNS) are a popular means of social communication for members of virtual communities - actors. At the same time, SNS have become an effective tool for conducting information operations directed against state information security. Therefore, an important scientific task is the timely detection of signs of information operations in the SNS. In the previous stages of research, a method for constructing profiles of information security actors in the SNS, which allows to assess the level of their threat as a possible participant in the information operation. The proposed method is generalized to all SNS and does not take into account the diversity of the set of attributes of profiles in individual services. Consequently, the perspective direction of research is the adaptation of this method for a specific SNS and its verification for further use in the system of providing information security of the state. An experimental study of the method is performed on the example of the microblogging Twitter. It is established that the accuracy and speed of the construction of profiles depends on the algorithm of the binary classification, which is used at the stage of assigning the actor to one of the given classes of threats. The obtained results coincide with the known academic studies, which testifies to the expediency of application of the developed method for automation of procedures for early detection of signs of information operations in the SNS.
References
2. Keenan A. Sociability and social interaction on social networking websites / A. Keenan, A. Shiri // LibraryReview. – Vol. 58, Iss. 6. – PP. 438–450.
3. Грищук Р. В. Основи кібернетичної безпеки : монографія / Р. В. Грищук, Ю. Г. Даник ; під заг. ред. Ю. Г. Даника. – Житомир : ЖНАЕУ, 2016. – 636 с.
4. Молодецька К. В. Узагальнена класифікація загроз інформаційній безпеці держави в соціальних інтернет-сервісах / К. В. Молодецька // Защита информации : сб. науч. труд. – 2016. – Вып. 23. – С. 75–87.
5. Определение демографических атрибутов пользователей микроблогов / А. Коршунов, И. Белобородов, А. Гомзин [и др.] // Труды Института системного программирования РАН. – 2013. – Т. 25. – С. 179–194.
6. Гомзин А. Г. Методы построения социо-демографических профилей пользователей сети Интернет / А. Г. Гомзин, С. Д. Кузнецов // Труды Института системного программирования РАН. – 2015. – Т. 27. – Вып. 4. – С. 129–143.
7. Pennacchiotti M. Democrats, republicans and Starbucks afficionados: user classification in Twitter / M. Pennacchiotti, A. M. Popescu // Proceedings of the 17th ACM SIGKDD international conference on Knowledge discovery and datamining. – ACM, 2011. – С. 430–438.
8. Beller C. I'm a Belieber: Social Roles via Self-identification and Conceptual Attributes / C. Belleretal // Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics. – 2014. – PP. 181–186.
9. Молодецька-Гринчук К. В. Метод побудови профілів інформаційної безпеки акторів соціальних інтернет-сервісів / К. В. Молодецька-Гринчук // Інформаційна безпека. – 2017. – № 2(26). – С. 104–110.
10. Горбулін В. П. Інформаційні операції та безпека суспільства: загрози, протидія, моделювання: монографія / В. П. Горбулін, О. Г. Додонов, Д. В. Ланде. – К. : Інтертехнологія, 2009. – 164 с.
11. MIB Datasets : [Online resource] / MIB Datasets. – Access mode : http://mib.projects.iit.cnr.it/dataset.html. – Title from the screen.
12. Weiss G. M. Learning when training data are costly: the effect of class distribution on tree induction / G. M. Weiss, F. Provost // Journal of Artificial Intelligence Research. – 2003. – 19. – PP. 315–354.
13. Weka 3 – Data Mining with Open Source Machine Learning Software in Java / Weka. – Access mode : http://www.cs.waikato.ac.nz/ml/weka/. – Title from the screen.
14. Меткалф Б. Закон Меткалфа сорок лет спустя после рождения Ethernet / Б. Меткалф // Открытые системы. СУБД. – 2014. – № 1. – С. 44–47.
15. Cresci S. Fame for sale: Efficient detection of fake Twitter followers / S. Cresci, R. Di Pietro, M. Petrocchi, A. Spognardi, M. Tesconi // Decision Support Systems. – 2015. – Vol. 80. – PP. 56–71.
16. Jensen U. Random Forest classification of Twitter users to detect features linked to bot susceptibility / U. Jensen, Chr. Schenk // Professional profile of Ulf Aslak. – Access mode : http://ulfaslak.com/portfolio/sigproc-sp.pdf. – Title from the screen.
11.09.2017
Downloads
-
PDF (Українська)
Downloads: 183
