期刊名称:International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
印刷版ISSN:2278-1323
出版年度:2014
卷号:3
期号:5
页码:1921-1925
出版社:Shri Pannalal Research Institute of Technolgy
摘要:The classification of uncertain data have a need to pay more attention in recent years due to the appearance of more and more database with uncertainties, such as sensor database, location database and biometric information systems. The objects in the uncertain database are vague and imprecise.Often we assume that data values are exact or precise in the database but data is sometimes inherently indecisive. There are some reasons owing to that errors are creep inside ¨C 1. Data obtained from physical device are often imprecise due to measurement errors 2. Quantization error introduced by the digitization process. 3. Applications like sensor network data values are continuously changing and recorded information is always stale 4. Indecision also comes from repeated measurements. Due to above reasons traditional data mining techniques cannot be applied directly on uncertain database. So there is a need to apply a novel technique that will be able to handle the uncertain database. In this paper we design a framework for classification of uncertain data using support vector machine and remove the problem in SVM using novel fuzzy C-means clustering algorithm (NFCC)
关键词:Uncertain data; SVM; Fuzzy C-means ; clustering; active and supervised learning; multy spectral data ; ; SMO