The Research On Incremental Evolution Characteristics Of Concept Cognitive Learning

As a new research field,concept cognitive learning analyzes the knowledge structure and related characteristics of data from a conceptual point of view.Among them,incremental concept cognitive learning has become a hot research topic,because it focuses on the process of concept learning and knowledge processing in the dynamic data environment,and conforms to the current situation of increasing data volume in the era of big data.This paper studies the causal asymmetry and stability of concept structure in the incremental process from the perspective of incremental evolution of concept cognitive learning.Firstly,based on the incremental concept tree,the visual analysis of causal asymmetry in incremental evolution of concept cognitive learning is proposed in this paper.Based on the attribute topology theory,this paper designed the model of incremental concept tree and its generation algorithm,then represented the asymmetry between the causes and results with different incremental concept tree structures in different incremental order according to the consistency of incremental concept tree and causal asymmetry in a time-dependent manner,thereby the visual analysis of causal asymmetry in the incremental evolution of concept cognitive learning is realized.Secondly,taking incremental concept tree as the research object,this paper proposed the stability analysis method of concept cognitive learning in incremental evolution process.In detailed,the structure similarity algorithm of incremental concept tree based on concept subtree,the node similarity algorithm of incremental concept tree based on concept importance,and the evaluation indexes are designed to calculate the similar of incremental concept tree,which are defined from macro and micro perspectives respectively.Moreover,combined with the above indicators,a global similarity algorithm which can dynamically adjust weight proportion according to different scenarios is designed.This paper analyzed the stability of incremental concept tree by evaluating the similarity of incremental concept tree between adjacent increments,so as to realize the trend analysis of stability in the incremental evolution process of concept cognitive learning.Finally,this paper takes the random formal context as the research object,and verifies the effectiveness of our method in analyzing incremental evolution characteristics of concept cognitive learning.The experiment presented the causal asymmetry and stability in the incremental evolution of concept cognitive learning,and compared it with the classical methods such as incremental concept lattice model and tree edit distance method.The experimental results show that the method presented in this paper is more prominent in analyzing incremental evolution of concept cognitive learning from the concept perspective,and it is more targeted in visual representation of incremental evolution characteristics of concept cognitive learning.Further,our method is more consistent with human cognitive process,and more universal for different application scenarios,which provides a new research perspective for the incremental evolution process of concept cognitive learning.