【摘要】 入侵检测存在不知道明确分类标准的分类问题,将聚类引入到入侵检测来改善入侵检测的性能是一个新的尝试。论文从应用聚类获取分类标准、利用聚类进行数据约简、聚类标记、相似性度量、警报聚类五个方面探讨了国内外在该领域内的研究成果,在此基础上进一步分析了存在的问题,并提了下一步的研究方向。更多还原

【Abstract】 Intrusion Detection is a classification problem which hasn’t the determined criterion. It is a new attempt to introduce the clustering technology into the IDS in order to improve its performance. In this paper, the latest research achievements in this field is reviewed, including such areas as determining classification criterion, data reduction with clustering, labeling clustering, similarity measurement and alarming clustering. Finally, the focuses for the future re- search are also presented.更多还原