匹配阈值的预测计算和自适应算法实现

基于否定选择算法，提出一种匹配阈值的预测计算方法，通过通用方程计算最优阈值，实验证明提高了检测率，同时分析了预测计算的适用范围。针对静态固定阈值的方法不能适应网络数据动态变化，利用遗传算法进化检测器同时实现阈值自适应，仿真结果表明提高了检测率，降低了误测率，能在自体集动态变化的情况下动态生成检测器。

Prediction calculation and self-adaptive algorithm for matching threshold in negative selection algorithm

Based on Negative Selection Algorithm (NSA), a prediction calculation strategy of matching threshold was put forward in this paper. Optimum value of matching threshold was achieved through general formula defined in this paper and simulation results prove that true positive was improved. The application scale of prediction strategy in matching rules was discussed. Since static matching threshold cannot meet the demand of dynamic network, self-adaptive matching threshold based on Genetic Algorithm (GA) was brought up for changeable network intrusion, which improved true positive and reduced false positive, and detectors were matured in dynamic under condition of changeable self set.