入侵檢測是檔案信息安全保障體系建設(shè)的重要組成部分，而在分布式入侵檢測系統(tǒng)
中，數(shù)據(jù)信息量大、格式非常多樣化，直接對原始數(shù)據(jù)進行處理和交換的效率非常低，針對此問題，采用一種基于元數(shù)據(jù)規(guī)范的方法，對數(shù)據(jù)對象進行規(guī)范化描述，統(tǒng)一數(shù)據(jù)格式，提高數(shù)據(jù)挖掘效率，為多系統(tǒng)、多模塊的協(xié)作提供了一種有效的數(shù)據(jù)交換方式。
關(guān)鍵詞：入侵檢測，元數(shù)據(jù)規(guī)范，數(shù)據(jù)挖掘，可擴展標識語言
Abstract: In Distributed Intrusion Detection System, it's inefficient if the object of data mining is raw data because of the large amount of information and diversified format. To solve this problem, a metadata standard method is adopted to standardize the description of metadata. This can improve the efficiency of the data mining, and provide an idea of the data exchange of multi-system and multi-module collaboration.
Key words: intrusion detection, metadata standards, data mining, extensible markup language