Title :
Research and improvement on ID3 algorithm in intrusion detection system
Author :
Zhai, Guangqun ; Liu, Chunyan
Author_Institution :
Sch. of Inf. Eng., Zhengzhou Univ., Zhengzhou, China
Abstract :
ID3 algorithm was a classic classification of data mining. It always selected the attribute with many values. The attribute with many values wasn´t the correct one, and it always created wrong classification. In the application of intrusion detection system, it would created fault alarm and omission alarm. To this fault, an improved decision tree algorithm was proposed. Though improvement of information gain formula, the correct attribute would be got. The decision tree was created after the data collected classified correctly. The tree would be not high and has a few of branches. The rule set would be got based on the decision tree. Experimental results showed the effectiveness of the algorithm, false alarm rate and omission rate decreased, increasing the detection rate and reducing the space consumption.
Keywords :
data mining; decision trees; security of data; ID3 algorithm; data mining; decision tree algorithm; intrusion detection system; Algorithm design and analysis; Classification algorithms; Classification tree analysis; Data analysis; Information entropy; Intrusion detection; ID3 algorithm; information entropy; information gain; intrusion detection; rule;
Conference_Titel :
Natural Computation (ICNC), 2010 Sixth International Conference on
Conference_Location :
Yantai, Shandong
Print_ISBN :
978-1-4244-5958-2
DOI :
10.1109/ICNC.2010.5582691
