在构造人工免疫系统中,如果能自动调节生成检测器的大小,从而利用较少的检测器,实现对较大“非自体集”的检测,就可从根本上提高系统效率。本文通过分析人工免疫系统中的主要算法——否定选择算法,以及检测器产生漏洞的原因,提出了一种可变阈值的检测器生成算法。与传统的否定选择算法相比,该算法大大减少了漏洞,使检测效率得到提高。
In this paper, an algorithm of generating the set of detectors with adjustable threshold of matching is proposed based on the analysis of the negative selection algorithm and the reason of appearing "holes". The new algorithm can reduce the number of holes and raise the rate of detecting.