论文部分内容阅读
认知计算是当今研究的热点,而研究表明形式背景可用于认知计算.形式背景有多种表示方式,属性拓扑作为形式背景的一种新型表示方法,能够直观的表现出属性之间的关联关系及关联强度.从认知角度出发,提出属性拓扑的激活方法与对应的物理意义.即在设定的形式背景中,通过属性拓扑构造形式背景形成的认知网络.当新对象出现时,根据新对象所包含的属性进行属性拓扑的网络顶点激活,并分析激活部分的子网络结构间的关联关系.通过对关联关系的分析与运算,完成新对象的识别与原有形式背景知识完备度的衡量与更新.实验数据表明,属性拓扑激活机制符合人类认知新事物过程,并揭示了以认知为基础的形式背景扩充必要条件.
Cognitive computing is a hot research topic nowadays, and research shows that formal context can be used in cognitive computing.Formal context has many kinds of representation, attribute topology as a new representation of formal context, can intuitively show the relationship between attributes Relationship and correlation intensity.From the cognitive perspective, we propose the activation method and the corresponding physical meaning of the attribute topology, that is, construct the cognitive network formed by the formal context through the attribute topology in the set formal context.When the new object appears, According to the attributes contained in the new object, the network vertex activation of the attribute topology is carried out, and the association between the sub-network structures in the active part is analyzed. Through the analysis and calculation of the relationship, the new object recognition and the completion of the original formal context knowledge Measurement and update.Experimental data show that attribute topological activation mechanism conforms to the process of new human cognition and reveals the necessary conditions for the extension of formal context based on cognition.