摘要
为提高否定选择算法中检测器集的检测率,提出改进的检测器集生成方法。其主要针对检测器在检测边界元素时遇到的困境问题,把自体点和它的临近点一起作为自体区域,处理自体的泛化问题。给出算法的具体实现过程、优势分析,并通过人工合成数据集2DSyntheticData和实际Biomedical数据集对算法进行了验证。实验结果表明,本算法检测率较高,尤其可以有效检测到处于自体与非自体边界处的点,具有一定的优越性。
In order to improve the detection rate of negative selection algorithm, proposed an improved detectors generation method. It aimed at solving the boundary dilemma problem. Regard the self and its neighboring as self regions. Given detailed realization and advantages of the algorithm. The experiments of synthetic and real data sets （ iris data set and biomedical data set） results show that the algorithm has higher detection rate, especially for the points in the boundary of self and nonself. So it has better performance.
出处
《计算机应用研究》
CSCD
北大核心
2011年第1期137-138,144,共3页
Application Research of Computers
关键词
否定选择算法
边界困境
检测器
检测率
negative selection algorithm
boundary dilemma
detector
detection rate
