Title :
A new uncertainty measure of knowledge in incomplete information systems
Author :
Li, Ren-Pu ; Huang, Dao
Author_Institution :
Inst. of Inf., East China Univ. of Sci. & Technol., Shanghai, China
Abstract :
In this paper, the problem of estimating the uncertainty of knowledge in incomplete information systems is studied. Some limitations of the previous measure - rough entropy are first analyzed and then a new measure called incomplete entropy is presented. Compared with rough entropy, incomplete entropy can be used in both incomplete and complete information system and has more precise estimation for uncertainty of knowledge in incomplete information systems.
Keywords :
entropy; estimation theory; information systems; knowledge based systems; rough set theory; uncertainty handling; incomplete entropy; incomplete information systems; knowledge uncertainty estimation; rough entropy analysis; Algorithm design and analysis; Cybernetics; Data mining; Entropy; Information analysis; Information systems; Machine learning; Measurement uncertainty; Set theory;
Conference_Titel :
Machine Learning and Cybernetics, 2004. Proceedings of 2004 International Conference on
Print_ISBN :
0-7803-8403-2
DOI :
10.1109/ICMLC.2004.1382008
