A Constructional Algorithm of Principal and Subordinate Structure Self-Set Based on Decision Tree

Aiming at solving the problem of "self-set incomplete" that exists in intrusion detection system based on computer immune, this paper designs a construction algorithm of principal and subordinate structure self-set based on decision trees. The decision trees are introduced to traditional negative-selection algorithm and the candidate detectors which have been eliminated by the immune tolerance are reclassified by the decision trees, and the candidate detectors that meet the setting conditions compose the "subordinate self-set" so as to achieve the dynamic expansion of the self-set. The unqualified elements in "subordinate self-set" are eliminated according to the "conflict of matching" method. Experimental results show that this algorithm is effective and can improve the recognition performance of the detectors.