论文信息 - Ignoring improper data in decision support system for medical applications

Ignoring improper data in decision support system for medical applications

A Decision Support System for Medical Applications was designed by applying the rough set theory to generate rules from the collected data. The data are kept in a table representing information system. There are some improper data in information systems and their removal can improve the quality of the retrieved information. By improper data we can understand such objects that disturb rules generation. They can be erroneous or corrupted or just exceptions. It is possible to find an algorithm of improper data removal to optimize the quality of information derived from decision tables. The improper data can be verified by checking whether some indicators of classification quality were improved after removal of the data. Some suggestions of identifying improper data are presented in the paper. In medical applications the improper data cannot be neglected.

[1] Andrzej Skowron,et al. EXTRACTING LAWS FROM DECISION TABLES: A ROUGH SET APPROACH , 1995, Comput. Intell..

[2] Zdzislaw Pawlak,et al. Rough classification , 1984, Int. J. Hum. Comput. Stud..

[3] J. Stefanowski,et al. Rough Sets as a Tool for Studying Attribute Dependencies in the Urinary Stones Treatment Data Set , 1997 .

[4] Roman Słowiński,et al. Derivation of optimal decision algorithms from decision tables using rough sets , 1988 .

[5] T. Y. Lin,et al. Rough Sets and Data Mining , 1997, Springer US.

[6] Ramin Yasdi,et al. Combining Rough Sets Learning- and Neural Learning-method to deal with uncertain and imprecise information , 1995, Neurocomputing.

[7] Zdzisław Pawlak,et al. Decision tables and decision algorithms , 1985 .