作者: Jianxiong Luo , Susan M. Bridges
DOI: 10.1002/1098-111X(200008)15:8<687::AID-INT1>3.0.CO;2-X
关键词:
摘要: Lee, Stolfo, and Mok 1 previously reported the use of association rules frequency episodes for mining audit data to gain knowledge intrusion detection. The integration with fuzzy logic can produce more abstract flexible patterns detection, since many quantitative features are involved in detection security itself is fuzzy. We present a modification algorithm rules, define concept episodes, an original episodes. add normalization step procedure order prevent one instance from contributing than others. also modify learn Experimental results show utility