| Dynamic concept learning is one of the active issues in the field of concept cognitive learning.According to the dynamic direction,it can be divided into incremental concept learning and decrement concept learning.Decrement concept learning focuses on the attenuation characteristics of concept structure in dynamic data.On the basis of dynamic concept learning,this paper studies graph attenuation and tree attenuation in the formal context of decrement form respectively from the perspective of attribute topology and concept tree.Firstly,this paper proposes an analysis method of attenuation characteristics of three-dimensional attribute topology.Attribute topology mainly describes the relationship between attributes,and the description of the strength of attributes has limitations.In order to describe the strength characteristics in the decay process of three-dimensional attribute topology,this paper extends the original definition of two-dimensional attribute topology and proposes the definition of three-dimensional attribute topology.According to the strength of attributes,the hierarchical representation of three-dimensional attribute topology is constructed.Combined with the strength characteristics of three-dimensional attribute topology,this paper analyzes the hierarchical attenuation of attributes over time,and proposes the attenuation properties of three-dimensional attribute topology and proves them.Secondly,this paper proposes a method to analyze the attenuation characteristics of concept tree.This paper analyzes the attenuation characteristics of the concept tree by combining the concept tree and the decrement problem in the concept cognitive learning.This paper first discusses the coupling of each node in the concept tree,and then proposes the corresponding node deletion algorithm for different types of nodes in the concept tree.Finally,according to the node coupling and node deletion algorithm,the concept tree is attenuated and the concept tree is updated,so as to complete the analysis of the attenuation characteristics of the concept tree.Finally,this paper takes the forgetting characteristics of human brain as the research object to verify the attenuation analysis method of the three-dimensional attribute topology and concept tree proposed in this paper.In the experiment,three-dimensional attribute topology and concept tree are used as models to simulate the process and results of human forgetting,and the results are compared with those of concept lattice method.The experimental results show that the scheme based on three-dimensional attribute topology and concept tree is feasible and reasonable.And it is proved that the research on the attenuation characteristics of dynamic concept learning has certain cognitive significance through experiments.In this paper,the methods of attenuation analysis of attribute topology and concept tree are proposed,which enrich the theory of attribute topology and concept tree,and deepen the research of attribute topology and concept tree in dynamic concept learning,and lay a foundation for its development and application in cognitive learning and artificial intelligence. |
