TY  - CONF
N1  - cited By 5
N2  - The objective of this research is to investigate the effects of missing attribute value imputation methods on the quality of extracted rules when rule filtering is applied. Three imputation methods: Artificial Neural Network with Rough Set Theory (ANNRST), k-Nearest Neighbor (k-NN) and Concept Most Common Attribute Value Filling (CMCF) are applied to University California Irvine (UCI) coronary heart disease data sets. Rough Set Theory (RST) method is used to generate the rules from the three imputed data sets. Support filtering is used to select the rules. Accuracy, coverage, sensitivity, specificity and Area Under Curve (AUC) of Receiver Operating Characteristics (ROC) analysis are used to evaluate the performance of the rules when they are applied to classify the complete testing data set. Evaluation results show that ANNRST is considered as the best method among k-NN and CMCF. © 2008 Springer-Verlag.
ID  - scholars495
SP  - 266
TI  - A comparative study of imputation methods to predict missing attribute values in coronary heart disease data set
KW  - Artificial heart; Biomedical engineering; Cardiology; Classification (of information); Diseases; Heart; Nearest neighbor search; Statistical tests
KW  -  Attribute values; Comparative studies; Coronary heart disease; imputation; Imputation methods; K-nearest neighbors; Receiver operating characteristics analysis; Rough set theory (RST)
KW  -  Rough set theory
AV  - none
A1  - Setiawan, N.A.
A1  - Venkatachalam, P.A.
A1  - Hani, A.F.M.
UR  - https://www.scopus.com/inward/record.uri?eid=2-s2.0-78349264542&doi=10.1007%2f978-3-540-69139-6_69&partnerID=40&md5=685440a65a1a064501d32e50350fc4ae
EP  - 269
VL  - 21 IFM
Y1  - 2008///
PB  - Springer Verlag
SN  - 16800737
ER  -