Knowledge Change Rate-Based Attribute Importance Measure and Its Performance Analysis

Chenxia Jin,Fachao Li,Qihui Hu
DOI: https://doi.org/10.1016/j.knosys.2016.12.002
IF: 8.139
2016-01-01
Knowledge-Based Systems
Abstract:Attribute importance measure is important in such approaches as data system reduction and, multi-attribute decisions. In this paper, we present knowledge change rate-based attribute importance measures with structural features of fuzzy measure, abbreviated as BCKCR–AIM. We discuss theoretical construction strategies and structural features followed by remarks on constructing BCKCR–AIM. Finally, experimental results for several examples and UCI data sets show the connections and differences between BCKCR–AIM and other attribute importance measures. The advantage of our measure is that it uses attributes set changes to describe knowledge change and associated features between lower and upper approximations of decision classes and knowledge to reflect attribute importance. Our measure can improve feasibility and interpretability; therefore, BCKCR–AIM has wide application in such approaches as attributes reduction, feature extraction, information fusion, and expert systems.
What problem does this paper attempt to address?