.png){kind=icon}
.png){kind=icon}
Return
Attribute importance measurement method based on data coordination degree
.png){kind=icon}
PRE
AI
DOI：10.1016/j.knosys.2019.105359.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The increasing scale of data information cause the great amount of irrelevant attributes, which becomes a challenging issue for machine learning. Therefore, removing the redundancy through sorting the attributes with appropriate significance has attracted wide attention in academic and application. Taking the knowledge hidden in the data system as the carrier and the inclusion relationship between sets as the basis, this paper proposes the concept of decision coordination degree. Then a composite attribute importance measurement based on core data is established (BCD-AICM). Further the basic properties and features of BCD-AICM are discussed. Finally, the similarities and differences between BCD-AICM and the existing attribute importance measurement methods are discussed using eight UCI data sets. The theoretical analysis and experiments results show that the BCD-AICM has good interpretability and structural characteristics. This method enriches the existing related theories and has broad application prospects in the fields of fuzzy decision-making, knowledge acquisition, resource management, and artificial intelligence etc. (C) 2019 Elsevier B.V. All rights reserved.
Keywords:
Data system
Core data set
Attribute importance
Decision coordination degree
Composite quantification
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
K
IF:
7.6
Papers:
1.2W
Citations:
4.5W
Organization
No organization information available