摘要：
The paper proposes a new network intrusion detection system based on fuzzy neural network by redesigning the intrusion detection system's architecture and arithmetic. In order to overcome the difficulty of specifying the membership function of rules depending on experiences of experts in multi-dimension space.,neural network is introduced to distinguish non-linearly input/output characteristics of complex system and to generate rule sets and membership functions automatically. The new architecture adopts the network processor to collect and analyse the data in the low layer of network.,and to establish a prototype system. This system demonstrated in this experiment appears to be better intrusion detection ability, moreover, which is able to detect unknown attack and plays down false alarms.
作者机构：
[王实; 张锦; 李睿] College of Software, Hunan Univ., Changsha 410082, China;[林雪梅] School of Information Science and Technology, Hunan Agricultural Univ., Changsha 410128, China;[成奋华] Dept. of Electron Information Engineering and Technology, Hunan Science Vocational College, Changsha 410082, China;[张锦] Dept. of Biomedical Engineering, Zhejiang Univ., Hangzhou 310027, China
作者机构：
[Tang, Hua; Zeng, Bi-Qing] Computer Engineering Department, South China Normal University, Foshan 528225, China;[Shen, Yue] School of Information Science Technology, Hunan Agricultural University, Changsha 410128, China
通讯机构：
Computer Engineering Department, South China Normal University, China
摘要：
With the wide application of intrusion detection systems (IDSs), the requirements of evaluating IDSs are getting urgent. This paper first summarizes the present status of IDSs evaluation. Based on the existing methods, an analytical method for decision tree analysis on cost for IDS evaluation is proposed. This method is based on ROC curve and introduces cost by decision tree. Then it optimizes the performance of IDSs with cost. Finally, the experiments are made to prove the validity with this method.
会议名称：
Parallel and Distributed Processing and Applications - ISPA 2005 Workshops, ISPA 2005 International Workshops AEPP, ASTD, BIOS, GCIC, IADS, MASN, SGCA, and WISA, Nanjing, China, November 2-5, 2005, Proceedings
摘要：
Along with high development of multimedia information technique, the provider of badness information embeds some badness information to image or directly saves as a image file, avoiding the filter of image, which brings extreme effect of security hidden trouble in society. An information audit system based on image content filtering is provided in this paper. At first, we discuss some basic method filtering physical badness image content, analyze some key technology of filtering image content, and mark as texture character by four eigenvectors: contrast, energy, entropy and correlation. Afterwards, we utilize dynamic programming method to segment image objects, and utilize similarity measurement to denote similarity degree of two character measures. At last, we give an example of identify yellow content, which distill the texture character of image and match it with defined character database. Our system can supervise and control badness information of physical badness image content, and realize automation audit of multimedia information.
作者机构：
[Jia, Yan; Wu, Quan-Yuan; Sui, Pin-Bo; Li, Ai-Ping] Sch. of Comp. Sci., Natl. Univ. of Defense Technol., Changsha 410073, China;[Liao, Gui-Ping] Dept. of Comp. and Info. Sci., Hu'nan Agric. Univ., Changsha 410073, China
通讯机构：
Sch. of Comp. Sci., Natl. Univ. of Defense Technol., China
关键词：
Diagnostic knowledge;Expert system;Knowledge acquisition;Knowledge representation;Rough set
摘要：
Knowledge acquisition is the bottleneck in developing expert system. It usually takes a long period to acquire disease knowledge using the traditional methods. Aiming at this problem, the paper presents the relationship between rough sets and rule-based rapeseed disease knowledge, namely the application of rough sets in knowledge acquisition from rapeseed disease expert. Then the exclusive rules, inclusive rules and disease images of rapeseed disease are built based on the RHINOS diagnosis model, and the definition of probability rule is put forward. At last, the paper presents the rule-based automated induction reasoning method, including exhaustive search, post-processing procedure, estimation for statistic test and the bootstrap and resampling methods. The results of experiment show that rough sets not only are a good framework for knowledge acquirement, but also can accurately induct the rules of plant diseases. This method can act as the assistant tool for development of diagnosis expert system, and has a extensive application in intelligent agriculture information systems.
