Title :
A Unified Paradigm for the Accuracy of Classification Based on Granular Computing
Author :
Chen, Yongbing ; Liu, Shuang ; Ye, Ping
Author_Institution :
Sch. of Math. & Inf. Sci., Zhejiang Normal Univ., Jinhua, China
Abstract :
Accuracy is a very important criterion for the classifier in the process of classification. In this paper, a unified paradigm for the calculation of accuracy evaluated different classifier, using topological covering-based granular computing, is presented under the given sample space and different ideal classification assumptions. And corresponding examples for the calculation of accuracy in different classification situations are given.
Keywords :
decision trees; pattern classification; accuracy calculation; ideal classification assumption; topological covering-based granular computing; unified paradigm; Accuracy; Conferences; Data mining; Estimation; Prediction methods; Rough sets; accuracy; classification; classifier; covering; granular computing;
Conference_Titel :
Granular Computing (GrC), 2010 IEEE International Conference on
Conference_Location :
San Jose, CA
Print_ISBN :
978-1-4244-7964-1
DOI :
10.1109/GrC.2010.34
