Detecting region outliers in meteorological data

  • Authors:
  • Jiang Zhao;Chang-Tien Lu;Yufeng Kou

  • Affiliations:
  • Virginia Polytechnic Institute and State University, Falls Church, VA;Virginia Polytechnic Institute and State University, Falls Church, VA;Virginia Polytechnic Institute and State University, Falls Church, VA

  • Venue:
  • GIS '03 Proceedings of the 11th ACM international symposium on Advances in geographic information systems
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

Spatial outliers are the spatial objects with distinct features from their surrounding neighbors. Detection of spatial outliers helps reveal important and valuable information from large spatial data sets. In the field of meteorology, for example, spatial outliers can be associated with disastrous natural events such as tornadoes, hurricane, and forest fires. Previous study of spatial outlier mainly focuses on point data. However, in the meteorological data or other applications, spatial outliers are frequently represented in region, i.e., a group of points, with two dimensions or even three dimensions, and the previous point-based approaches may not be appropriate to be used. As region outliers are commonly multi-scale objects, wavelet analysis is an effective tool to study them. In this paper, we propose a wavelet analysis based approach to detect region outliers. We discuss the region outlier detection problem and design a suite of algorithms to effectively discover them. The algorithms were implemented and evaluated with a real-world meteorological data set.