摘要
提出一个动态可增殖的多层次自组织认知系统,每个层次具有形式上一致的知识表示方法,各层的自组关联、自组聚合、归约和样本表达四个知识处理模型是实现系统自组织层次增殖的核心模型。指出若要实现层次可自组织增殖的系统,其关键是要设计一个合理的聚合归约演算系统;提出一个适用于各个层次的基于可结合半边的自组图知识表示法,先给出自组图形式化的静态定义和动态定义,然后以自组关联模型为背景给出对应的自组图构造算法。
A generality model for dynamical grow recognition system with hierarchical self-organization has been proposed.There has a formal unity knowledge representation in every layer.There are four knowledge process models for each layer,the first model is self-organizing connection model,the second is self-organizing aggregation model,the third is reduction model and the last is sample representation model.These four models are the core algorithms for the dynamical grow recognition system with hierarchical self-organization.This paper points out,to design a reasonable aggrega- tion and reduction calculus model is the critical step for the recognition system that can grow with hierarchical self-or- ganization.For this purpose,a new knowledge representation method called self-organization graph has been proposed.The self-organization graph is a formal calculus model based on mated half edge;the half edge is also a new concept in graph calculus model.The mated half edge is inspired from the supramolecular self-assembly and supramolecular selfrecognition.The static definition for self-organization graph is given.To realize the first knowledge process model i.e.the self-organizing connection model,propose a dynamic definition for self-organization graph.Lastly,a graph generate algorithm is given.
出处
《计算机工程与应用》
CSCD
北大核心
2006年第29期43-48,51,共7页
Computer Engineering and Applications
关键词
多层次自组织
构造演算
半边
自组图
hierarchical self-organization,generation calculus,half edge,self-organization graph
