文章基本信息

标题：Feature Selection with Support Vector Machines Applied on Tornado Detection
其他标题：Feature Selection with Support Vector Machines Applied on Tornado Detection
本地全文：下载
作者：Budi Santosa
期刊名称：Majalah Iptek = IPTEK : The Journal for Technology and Science
印刷版ISSN：0853-4098
电子版ISSN：2088-2033
出版年度：2007
卷号：18
期号：1
DOI：10.12962/j20882033.v18i1.178
语种：English
出版社：IPTEK
摘要：In this paper, a linear programming support vector machine which is based on L1-norm is applied to do feature selection in the tornado data set. The data is the ouputs of Weather Surveillance Radar 1998 Doppler (WSR-88D). The approach is evaluated based on the indices of probability of detection, false alarm rate, bias and Heidke skill. Tornado circulation attributes/variables derived largely from the National Severe Storms Laboratory Mesocyclone Detection Algorithm (MDA) have been investigated for their efficacy in distinguishing between mesocyclones that become tornadic from those which do not.
关键词：Classification; Detection; Feature Selection; Bayesian Neural Networks; Machine Learning; Linear Programming Support Vector Machines; Linear Discriminant Analysis; Performance Indices