GeGeM:一种人工免疫系统通用基因模型及实现

GeGeM:General Gene Model for Artificial Immune System with Implementation

一个通用的基因模型对于人工免疫系统AIS(ArtificialImmuneSystem)软件设计是必需的。从AIS软件架构设计的角度出发,基因模型应具有一般性、可扩展性、高效率和可用性。提出一个基因模型GeGeM(GeneralGeneModel),通过基因操作实现免疫计算。该模型基于三层结构:元基因、基因序列和基因数据集。其中元基因提供基因序列的规范,基因序列提供基本的公共的基因操作,而基因数据集在此基础上实现训练和检测。进一步,该模型实现并给出实验结果,结果分析表明该模型的可用性和算法的有效性。经讨论比较,该模型具有一般性和可扩展性,可用于建立多用途的AIS软件,也可用于构建特定领域的复杂多样的检测系统。

A general gene model is necessary for AIS （Artificial Immune System） software design. From the context of AIS software architecture, a gene model should be general, extendable, efficient and available. A General Gene Model （GeGeM） for AIS to support immune computing was proposed for gene operating. The model is designed based on three layers： Meta-Gene, Gene-Serial and Gene-Dataset. The Meta-Gene layer specifies the type of Gene-Serial. Gene-Serial provides basic general operations on genes, while Gene-Dataset realizes training and detecting based on Gene-Serial. Further, the model gets implemented and experimental results indicate that the GeGeM model is available and effective. It is discussed and compared that the model is general and extensible. It can be employed to build multi-purpose AIS software, and further, it is expected to be utilized to construct complex and diverse detection systems for particular fields .