结果可控树突状细胞算法  

RESULT-CONTROLLED DENDRITIC CELL ALGORITHM

作　　者：张贺[1] 袁嵩[1] 

机构地区：[1]武汉科技大学计算机科学与技术学院,湖北武汉430065

出　　处：《计算机应用与软件》2013年第1期59-62,74,共5页Computer Applications and Software

基　　金：国家自然科学基金项目(60975031)

摘　　要：为解决传统树突状细胞算法因涉及大量不确定因素而导致的难以协调误报率和漏报率的关系的问题并在传统树突状细胞算法的基础上提高算法的检测精度,通过对树突状细胞算法信号转换公式、权值矩阵及其他不确定参数的分析,提出倾向因子的概念和结果可控树突状细胞算法。该算法具有更高的检测精度和更好的普适性,能够根据不同的细胞环境确定相应的倾向因子,进而实现对检测结果的调控能力。通过使用不同参数和多种数据集进行仿真实验,证明了概念的合理性和算法的有效性。To coordinate the relationship between false positive rate and false negative rate-of which the classical dendritic cell algorithm is difficult to analyse due to a great deal of probabilistie elements, and to improve the detection accuracy based on the classical algorithm, we propose the concept of tendency factor and the result-controllable dendritic cell algorithm through analysing the signal conversion formula, weighting matrix and the other random-based parameters involved in dendritic cell algorithm. The new algorithm has higher detection accuracy and better universality. According to different cell environment, the corresponding tendency factor can be calculated to realise the control ca- pability on detection results. Simulative experiments are performed using different parameters and multiple data sets, the reasonability of the concept and the＇ effectiveness of the algorithm are proved.

关 键 词：树突状细胞算法 倾向因子 危险理论 人工免疫 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]