Open AccessIncremental Attribute Reduction Based on Knowledge Granularity under Incomplete Data
Author(s) -
Qiangqiang Zhong,
Lei Wang,
Wen Yang,
Chao Liu
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/2025/1/012042
Subject(s) - granularity , data mining , reduction (mathematics) , object (grammar) , computer science , matrix (chemical analysis) , attribute domain , focus (optics) , simple (philosophy) , measure (data warehouse) , rough set , algorithm , artificial intelligence , mathematics , philosophy , materials science , geometry , physics , epistemology , optics , composite material , operating system
Traditional attribute reductions can’t be directly applied to miss attribute values and dynamic changes data. Therefore, how to dynamically update attribute reductions efficiently when incomplete data changes dynamically is great research value. For the change of the object, using matrix’s simple and clear characteristics, the paper focus on the incremental attribute reduction algorithm in matrix form of knowledge granularity. Firstly, the knowledge granularity is obtained by matrix which uses to measure the uncertainty of the system, then calculating the attribute importance, and using the attribute importance to devise a static algorithm, and then analyzing incremental matrix form when the object increases. The comparative experiments show that the incremental algorithm can effectively and efficiently update the result of attribute reduction as the object changes.