Home // INTELLI 2016, The Fifth International Conference on Intelligent Systems and Applications // View article

Bagged Extended Nearest Neighbors Classification for Anomalous Propagation Echo Detection

Authors:
Hansoo Lee
Hye-Young Han
Sungshin Kim

Keywords: Extended nearest neighbors; Hamamoto’s bootstrap II; Anomalous propagation echo; Weather prediction; Classification.

Abstract:
Abstract—Radar is one of essential and popular devices in weather prediction process because of its wide array of advantages. Unfortunately, the observation results contains lots of unwanted radar signals and they disrupt forecasting process. The representative non-precipitation echoes are permanent, spurious, and anomalous propagation echoes. Among them, the anomalous propagation echo can be a source of severely negative influences in a quantitative precipitation estimation. Therefore, a reliable automatic systems for identifying the anomalous propagation echo is needed. In this paper, we suggest a novel k-nearest neighbors algorithm, by combining the Hamamoto’s bootstrap II method and the extended nearest neighbors for improving performance of the classifier. Using the actual appearance cases of the anomalous propagation echo, it is confirmed that the suggested method is better than the k-nearest neighbors and the extended nearest neighbors.

Pages: 39 to 44

Copyright: Copyright (c) IARIA, 2016

Publication date: November 13, 2016

Published in: conference

ISSN: 2308-4065

ISBN: 978-1-61208-518-0

Location: Barcelona, Spain

Dates: from November 13, 2016 to November 17, 2016