CONDENSER: A Graph-Based Approachfor Detecting Botnets
نویسندگان
چکیده
Botnets represent a global problem and are responsible for causing large financial and operational damage to their victims. They are implemented with evasion in mind, and aim at hiding their architecture and authors, making them difficult to detect in general. These kinds of networks are mainly used for identity theft, virtual extortion, spam campaigns and malware dissemination. Botnets have a great potential in warfare and terrorist activities, making it of utmost importance to take action against. We present CONDENSER, a method for identifying data generated by botnet activity. We start by selecting appropriate the features from several data feeds, namely DNS non-existent domain responses and live communication packages directed to command and control servers that we previously sinkholed. By using machine learning algorithms and a graph based representation of data, then allows one to identify botnet activity, helps identifying anomalous traffic, quickly detect new botnets and improve activities of tracking known botnets. Our main contributions are threefold: first, the use of a machine learning classifier for classifying domain names as being generated by domain generation algorithms (DGA); second, a clustering algorithm using the set of selected features that groups network communication with similar patterns; third, a graph based knowledge representation framework where we store processed data, allowing us to perform queries.
منابع مشابه
Detecting Active Bot Networks Based on DNS Traffic Analysis
Abstract—One of the serious threats to cyberspace is the Bot networks or Botnets. Bots are malicious software that acts as a network and allows hackers to remotely manage and control infected computer victims. Given the fact that DNS is one of the most common protocols in the network and is essential for the proper functioning of the network, it is very useful for monitoring, detecting and redu...
متن کاملDNS Traffic Analysis for Network-based Malware Detection
(English) Botnets are generally recognized as one of the most challenging threats on the Internet today. Botnets have been involved in many attacks targeting multinational organizations and even nationwide internet services. As more effective detection and mitigation approaches are proposed by security researchers, botnet developers are employing new techniques for evasion. It is not surprising...
متن کاملA graph-theoretic framework for isolating botnets in a network
We present a new graph-based approach for the detection and isolation of botnets in a computer network. Our approach depends primarily on the temporal co-occurrences of malicious activities across the computers in a network and is independent of botnet architectures and the means used for their command and control. As practically all aspects of how a botnet manifests itself in a network—such as...
متن کاملA Novel Approach for Detecting Relationships in Social Networks Using Cellular Automata Based Graph Coloring
All the social networks can be modeled as a graph, where each roles as vertex and each relationroles as an edge. The graph can be show as G = [V;E], where V is the set of vertices and E is theset of edges. All social networks can be segmented to K groups, where there are members in eachgroup with same features. In each group each person knows other individuals and is in touch ...
متن کاملBotXrayer : Exposing Botnets by Visualizing DNS Traffic
Botnets pose a major problem to Internet security. They can cause various online crimes such as DDoS attacks, identity thefts and spam e-mails. While there have been many attempts to detect botnets, most of these studies have difficulties in detecting botnets due to their evasive techniques to resemble normal traffic. In this paper, we propose a visualization method, BotXrayer, to detect botnet...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید
ثبت ناماگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید
ورودعنوان ژورنال:
- CoRR
دوره abs/1410.8747 شماره
صفحات -
تاریخ انتشار 2014