We show that there exist two behaviorally distinct categories of spammers and that they employ different spamming strategies. Our analysis includes over 100 million messages collected from Twitter over the course of one month. In this work, we present a unique analysis of spam accounts in OSNs viewed through the lens of their behavioral characteristics. As spammers continuously keep creating newer accounts and evasive techniques upon being caught, a deeper understanding of their spamming strategies is vital to the design of future social media defense mechanisms. Spam in Online Social Networks (OSNs) is a sys-temic problem that imposes a threat to these services in terms of undermining their value to advertisers and potential investors, as well as negatively affecting users' engagement. We further provide a mapping between the proposed framework and National Institute of Standards and Technology (NIST) 800-30 risk assessment model.
Finally, in the clustering stage, we propose Streaming K-means algorithm to detect campaign of spam messages. We define a risk assessment function that computes the risk score from the outputs of Multinomial NB and KNN algorithms. The proposed framework is based on data stream classification and clustering, where the risk associated with a message is determined using a combination of Multinomial Naive Bayes (Multinomial NB) and modified K-nearest neighbor (KNN) algorithms. Our framework is inspired by the need to assess the severity level of microblogging messages in real-time in order to cope with the large volume of data generated on a daily basis. In this paper, we propose an ensemble based streaming framework for spam detection and risk assessment. Existing spam detection models developed for microblogging social network focused on batch learning approach and they classified messages as either spam or legitimate. These distinguished characteristics present challenges to traditional email and webpage spam detection filters. Unlike other social networking sites, such as Facebook and LinkedIn, which allow users to post messages with long characters, microblogging post has distinguished features: (1) Microblogging messages are short, composed within a limited available character length (2) The messages contain many noisy data and are very unstructured in nature (3) Microblogging messages include many domain specific words. Microblogging social network, like Twitter, Sina Weibo and Tumblr, has become attractive platform for social spammers to disseminate malicious contents. To the best of our knowledge, this is the first survey to discuss DNS-based botnet detection techniques in which the problems, existing solutions and the future research direction in the field of botnet detection based on DNS traffic analysis for effective botnet detection mechanisms in the future are explored and clarified. Therefore, this paper comes up to explore the various botnet detection techniques through providing a survey to observe the current state of the art in the field of botnet detection techniques based on DNS traffic analysis. Fortunately, different approaches have been proposed and developed to tackle the problem of botnets however, the problem still rises and emerges causing serious threat to the cyberspace-based businesses and individuals. It is also worth mentioning that the current trend of botnets is to hide their identities (i.e., the command and control server) using the DNS services to hinder their identification process.
#INVESTIGATE AN ANOMALY DETECTED NEAR CATTY CORNER SOFTWARE#
Botnet is a group of compromised hosts running malicious software program for malicious purposes, known as bots.
Botnet is a thorny and a grave problem of today’s Internet, resulting in economic damage for organizations and individuals.
0 kommentar(er)
