Taekyong Nam

Implementation of high performance objectionable video classification system

Sex video files being spread via Web site and P2P networks have caused social problems. It is necessary to prevent more people, especially teenagers, from getting a distorted view of sex through the influence of the obscene content. We propose and implement the obscene video classification system using two visual features to determine a videos lewdness with very excellent performance. In our approach, the two features are first built for each video and a final decision function is made to classify an obscene video. One of the visual features is a single frame based decision variable and the other is a group frame based decision variable. Then the final decision function is developed to optimize high classification performance using the variables by the discriminant analysis. Experimentally we show that the proposed method performs excellently in classifying videos into porno and the other, thus enabling automatic filtering of obscene videos

An efficient text filter for adult Web documents

The openness of the Web allows any users to access almost any type of information. However, some information, such as adult content, is not appropriate for all users, notably children. Additionally for adults, some contents included in abnormal pornographic sites can do ordinary peoples mental health harm. In this paper, we propose a new criterion and divide contents of Web documents into 4 grades. We use a hierarchical way of filtering texts. At first, we filter off 0-grade texts contain no adult contents using a pattern matching algorithm, and classify 1-grade, 2-grade and 3-grade texts using a machine learning algorithm

P2P Honeypot to Prevent Illegal or Harmful Contents From Spreading in P2P network

In this paper we propose the P2P Honeypot that prevents illegal or harmful files from spreading in P2P network. We apply the idea of Honeypot to P2P network. We build fake P2P service farms and monitor and trace users who spread or gain illegal or harmful files in P2P network. If we can apply this system widely, we can expect illegal or harmful files to be wiped out in P2P network.

A comparative study of the objectionable video classification approaches using single and group frame features

This paper deals with the methods for classifying whether a video is harmful or not and also evaluates their performance. The objectionable video classification can be performed using two methods. One can be practiced by judging whether each frame included in the video is harmful, and the other be obtained by using the features reflecting the entire characteristics of the video. The former is a single frame-based feature and the latter is a group frame-based feature. Experimental results show that the group frame-based feature outperforms the single frame-based feature and is robust to the objectionable video classification.

A harmful content protection in peer-to-peer networks

A variety of activities like commerce, education, voting is taking place in cyber-space these days. As the use of the Internet gradually increases, many side effects have appeared. Therefore, it is time to address the many troubling side effects associated with the Internet. In this paper, we propose a method of prevention of adult video and text in Peer-to-Peer networks and we evaluate text categorization algorithms for Peer-to-Peer networks. We also suggest a new image categorization algorithm for detecting adult images and compare its performance with existing commercial software.

2-Way text classification for harmful web documents

The openness of the Web allows any user to access almost any type of information. However, some information, such as adult content, is not appropriate for all users, notably children. Additionally for adults, some contents included in abnormal porn sites can do ordinary peoples mental health harm. In this paper, we propose an efficient 2-way text filter for blocking harmful web documents and also present a new criterion for clear classification. It filters off 0-grade web texts containing no harmful words using pattern matching with harmful words dictionaries, and classifies 1-grade,2-grade and 3-grade web texts using a machine learning algorithm.

The expansion of key infection model for dynamic sensor network

The establishment of shared cryptography keys is one of the challenging problems in the sensor networks. Key infection (R. Anderson et al., 2004) is a promising model to solve this problem without complex mechanism on the commodity sensor networks. This model, however, does not consider the mobility of sensor, so it can not support dynamic sensor network fields. Therefore, key infection model has to be extended to handle the mobility of sensor, and then an extended key infection model can be used on the dynamic sensor network. In this paper, we propose a scheme to extend the key infection model for supporting dynamic sensor networks and explain how the proposed scheme is operated in detail. Also we prove that the proposed scheme is secure

A Distinction Technology for Harmful Web Documents by Rates

The openness of the Web allows any user to access almost any type of information easily at any time and anywhere. However, with function of easy access for useful information, internet has dysfunctions of providing users with harmful contents indiscriminately. Some information, such as adult content, is not appropriate for all users, notably children. Additionally for adults, some contents included in abnormal porn sites can do ordinary people`s mental health harm. In the meantime, since Internet is a worldwide open network it has a limit to regulate users providing harmful contents through each countrie`s national laws or systems. Additionally it is not a desirable way of developing a certain system-specific classification technology for harmful contents, because internet users can contact with them in diverse way, for example, porn sites, harmful spams, or peer-to-peer networks, etc. Therefore, it is being emphasized to research and develop context-based core technologies for classifying harmful contents. In this paper, we propose an efficient text filter for blocking harmful texts of web documents using context-based technologies.

High-Speed concurrent attack responding system using FPGA supporting IPv4/IPv6

Recently, ISP needs more accurate and high-speed attack prevention system to manage the network more safely and clean. Securing the network, host, and on-line application is absolutely important. At the same time, the security function must not disturb productivity. To ensure that increasing network traffic is safe and their networks are secure, these organizations must provide security with bias toward solutions that accommodate performance demands, while providing the security and networking features required to run their businesses. That is, best solutions are those that combine high performance with topnotch security. For satisfying those requirements, we have developed hardware based and high performance ATPS (adaptive threat prevention system) which provides security functions such as IPS, IDS, rate-limiting, and traffic metering in wire speed. ATPS can manage and inspect IPv4/IPv6 packet in dual-stack mode.

On the Security Effect of Abnormal Traffic Controller Deployed in Internet Access Point

ATC (Abnormal traffic controller) is presented as next generation security technology to securely support reliable Internet service in traffic-intended unknown attack. The key concept of the ATC is abnormal traffic monitoring and traffic control technology. When fault factors exist continuously and/or are repeated, abnormal traffic control guarantees service completeness as much as possible. The ATC with control policy on abnormal traffic is superior to the ATC with blocking policy as well as conventional network node, when the ratio of effective traffic to abnormal traffic is higher than 30%. As the proposed ATC can be applied to the edge point of backbone network as well as Internet access point, it guarantees network survivability and provides reliable Internet service.