Attribute Selection Based on FRiS-Compactness

Nikolay Zagoruiko, Irina Borisova, Vladimir Dyubanov, Olga Kutnenko
Proceedings of the Fourth International Workshop on Feature Selection in Data Mining, PMLR 10:35-44, 2010.

Abstract

Commonly to classify new object in Data Mining one should estimate its similarity with given classes. Function of Rival Similarity (FRiS) is assigned to calculate quantitative measure of similarity considering a competitive situation. FRiS-function allows constructing new effective algorithms for various Data Mining tasks solving. In particular, it enables to obtain quantitative estimation of compactness of patterns which can be used as indirect criterion for informative attributes selection. FRiS-compactness predicts reliability of recognition of control sample more precisely, than such widespread methods as One-Leave-Out and Cross-Validation. Presented in the paper results of real genetic task solving confirm efficiency of FRiS-function using in attributes selection and decision rules construction.

Cite this Paper


BibTeX
@InProceedings{pmlr-v10-zagoruiko10a, title = {Attribute Selection Based on FRiS-Compactness}, author = {Zagoruiko, Nikolay and Borisova, Irina and Dyubanov, Vladimir and Kutnenko, Olga}, booktitle = {Proceedings of the Fourth International Workshop on Feature Selection in Data Mining}, pages = {35--44}, year = {2010}, editor = {Liu, Huan and Motoda, Hiroshi and Setiono, Rudy and Zhao, Zheng}, volume = {10}, series = {Proceedings of Machine Learning Research}, address = {Hyderabad, India}, month = {21 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v10/zagoruiko10a/zagoruiko10a.pdf}, url = {https://proceedings.mlr.press/v10/zagoruiko10a.html}, abstract = {Commonly to classify new object in Data Mining one should estimate its similarity with given classes. Function of Rival Similarity (FRiS) is assigned to calculate quantitative measure of similarity considering a competitive situation. FRiS-function allows constructing new effective algorithms for various Data Mining tasks solving. In particular, it enables to obtain quantitative estimation of compactness of patterns which can be used as indirect criterion for informative attributes selection. FRiS-compactness predicts reliability of recognition of control sample more precisely, than such widespread methods as One-Leave-Out and Cross-Validation. Presented in the paper results of real genetic task solving confirm efficiency of FRiS-function using in attributes selection and decision rules construction.} }
Endnote
%0 Conference Paper %T Attribute Selection Based on FRiS-Compactness %A Nikolay Zagoruiko %A Irina Borisova %A Vladimir Dyubanov %A Olga Kutnenko %B Proceedings of the Fourth International Workshop on Feature Selection in Data Mining %C Proceedings of Machine Learning Research %D 2010 %E Huan Liu %E Hiroshi Motoda %E Rudy Setiono %E Zheng Zhao %F pmlr-v10-zagoruiko10a %I PMLR %P 35--44 %U https://proceedings.mlr.press/v10/zagoruiko10a.html %V 10 %X Commonly to classify new object in Data Mining one should estimate its similarity with given classes. Function of Rival Similarity (FRiS) is assigned to calculate quantitative measure of similarity considering a competitive situation. FRiS-function allows constructing new effective algorithms for various Data Mining tasks solving. In particular, it enables to obtain quantitative estimation of compactness of patterns which can be used as indirect criterion for informative attributes selection. FRiS-compactness predicts reliability of recognition of control sample more precisely, than such widespread methods as One-Leave-Out and Cross-Validation. Presented in the paper results of real genetic task solving confirm efficiency of FRiS-function using in attributes selection and decision rules construction.
RIS
TY - CPAPER TI - Attribute Selection Based on FRiS-Compactness AU - Nikolay Zagoruiko AU - Irina Borisova AU - Vladimir Dyubanov AU - Olga Kutnenko BT - Proceedings of the Fourth International Workshop on Feature Selection in Data Mining DA - 2010/05/26 ED - Huan Liu ED - Hiroshi Motoda ED - Rudy Setiono ED - Zheng Zhao ID - pmlr-v10-zagoruiko10a PB - PMLR DP - Proceedings of Machine Learning Research VL - 10 SP - 35 EP - 44 L1 - http://proceedings.mlr.press/v10/zagoruiko10a/zagoruiko10a.pdf UR - https://proceedings.mlr.press/v10/zagoruiko10a.html AB - Commonly to classify new object in Data Mining one should estimate its similarity with given classes. Function of Rival Similarity (FRiS) is assigned to calculate quantitative measure of similarity considering a competitive situation. FRiS-function allows constructing new effective algorithms for various Data Mining tasks solving. In particular, it enables to obtain quantitative estimation of compactness of patterns which can be used as indirect criterion for informative attributes selection. FRiS-compactness predicts reliability of recognition of control sample more precisely, than such widespread methods as One-Leave-Out and Cross-Validation. Presented in the paper results of real genetic task solving confirm efficiency of FRiS-function using in attributes selection and decision rules construction. ER -
APA
Zagoruiko, N., Borisova, I., Dyubanov, V. & Kutnenko, O.. (2010). Attribute Selection Based on FRiS-Compactness. Proceedings of the Fourth International Workshop on Feature Selection in Data Mining, in Proceedings of Machine Learning Research 10:35-44 Available from https://proceedings.mlr.press/v10/zagoruiko10a.html.

Related Material