Abstract : Networking, which is one of the most significant aspects of information technology revolution, is developing increasingly day after day. This is because it offers a huge amount of knowledge, resources and human experiences. On the one hand, it contains a considerable amount of harmful content, because of misusing. On the other hand, sitting for a long time in front of PC’s or other network-based devices can affect body badly. As enterprise computing environments become more network-oriented, the importance of network traffic monitoring and analysis intensifies. Most existing traffic monitoring and analysis tools focus on measuring the traffic loads of individual network segments. Further, they typically have complicated user interfaces. This paper introduces and presents the design an application and implementation of an MS Windows-compatible software tool that is used to manage networks usage and keep track of every network user activity. An application consists of two parts client and server. The client side is a backgroundapplication runs whenever the PC is run, it turns off only when the PC is turned off and launched with its startup. The server side is more complex-GUI application that is responsible mainly for receiving data sent by clients group, managing and updating data to provide network owner up to date view. The effectiveness of an application has been verified by applying it to an enterprise network environment.
? Most existing traffic monitoring and analysis tools focus on measuring the traffic loads of individual network segments.
? The system uses exist networking protocols to communicate with router, and other devices and control them remotely.
? Traffic measurement and analysis are crucial to the design, operation and maintenance of wide-area networks based on Internet Protocol.
? This measurement data is used for traffic engineering, performance debugging, network operations, and to measure compliance with performance targets.
? They require additional modern network traffic analysis tools in order to manage network, solve the network problems quickly to avoid network failure, and handle the network security.
? The decision tree (DT) is powerful and popular supervised machine learning for decision-making and classification problems.
? Neural networks provide a solution to the problem of network traffic prediction by applying numbers of techniques.
• This paper presents a review of several techniques proposed, used and practiced for network traffic analysis and prediction.
• The graph based clustering algorithm is proposed for partition data set into a number of clusters. They proposed rough set theory to extract the relevant attributes from entire data set and for classification.
• The objective of the proposed project here that it is designed an application to help networks owners to control their networks in a proper way by providing them necessary data and controlling permissions over their clients.
? The sensitivity and false positive rate metrics are applied to measure the performances of algorithms.
? A number of current studies have reported that the SVM results of giving higher performance with respect to classification accuracy than the other classification approaches.
? The detection rate metric is presented to measure the performance of rough set theory approach.
? They used positive rate, accuracy metrics at the training and testing time to measure performance of their approaches.
|