Applying a machine learning technique to classification of Japanese pressure patterns

H. Kimura, Hideyuki Kawashima, H. Kusaka, H. Kitagawa

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

In climate research, pressure patterns are often very important. When a climatologists need to know the days of a specific pressure pattern, for example "low pressure in Western areas of Japan and high pressure in Eastern areas of Japan (Japanese winter-type weather)," they have to visually check a huge number of surface weather charts. To overcome this problem, we propose an automatic classification system using a support vector machine (SVM), which is a machine-learning method. We attempted to classify pressure patterns into two classes: "winter type" and "non-winter type". For both training datasets and test datasets, we used the JRA-25 dataset from 1981 to 2000. An experimental evaluation showed that our method obtained a greater than 0.8 F-measure. We noted that variations in results were based on differences in training datasets.

Original languageEnglish
JournalData Science Journal
Volume8
DOIs
Publication statusPublished - 2009 Mar 30
Externally publishedYes

Fingerprint

Learning systems
Support vector machines

Keywords

  • Classification
  • Machine learning
  • Pressure pattern
  • Support vector machine (SVM)

ASJC Scopus subject areas

  • Computer Science (miscellaneous)
  • Computer Science Applications

Cite this

Applying a machine learning technique to classification of Japanese pressure patterns. / Kimura, H.; Kawashima, Hideyuki; Kusaka, H.; Kitagawa, H.

In: Data Science Journal, Vol. 8, 30.03.2009.

Research output: Contribution to journalArticle

@article{f33a4d2175c54429869666486f425ec4,
title = "Applying a machine learning technique to classification of Japanese pressure patterns",
abstract = "In climate research, pressure patterns are often very important. When a climatologists need to know the days of a specific pressure pattern, for example {"}low pressure in Western areas of Japan and high pressure in Eastern areas of Japan (Japanese winter-type weather),{"} they have to visually check a huge number of surface weather charts. To overcome this problem, we propose an automatic classification system using a support vector machine (SVM), which is a machine-learning method. We attempted to classify pressure patterns into two classes: {"}winter type{"} and {"}non-winter type{"}. For both training datasets and test datasets, we used the JRA-25 dataset from 1981 to 2000. An experimental evaluation showed that our method obtained a greater than 0.8 F-measure. We noted that variations in results were based on differences in training datasets.",
keywords = "Classification, Machine learning, Pressure pattern, Support vector machine (SVM)",
author = "H. Kimura and Hideyuki Kawashima and H. Kusaka and H. Kitagawa",
year = "2009",
month = "3",
day = "30",
doi = "10.2481/dsj.8.S59",
language = "English",
volume = "8",
journal = "Data Science Journal",
issn = "1683-1470",
publisher = "Committee on Data for Science and Technology",

}

TY - JOUR

T1 - Applying a machine learning technique to classification of Japanese pressure patterns

AU - Kimura, H.

AU - Kawashima, Hideyuki

AU - Kusaka, H.

AU - Kitagawa, H.

PY - 2009/3/30

Y1 - 2009/3/30

N2 - In climate research, pressure patterns are often very important. When a climatologists need to know the days of a specific pressure pattern, for example "low pressure in Western areas of Japan and high pressure in Eastern areas of Japan (Japanese winter-type weather)," they have to visually check a huge number of surface weather charts. To overcome this problem, we propose an automatic classification system using a support vector machine (SVM), which is a machine-learning method. We attempted to classify pressure patterns into two classes: "winter type" and "non-winter type". For both training datasets and test datasets, we used the JRA-25 dataset from 1981 to 2000. An experimental evaluation showed that our method obtained a greater than 0.8 F-measure. We noted that variations in results were based on differences in training datasets.

AB - In climate research, pressure patterns are often very important. When a climatologists need to know the days of a specific pressure pattern, for example "low pressure in Western areas of Japan and high pressure in Eastern areas of Japan (Japanese winter-type weather)," they have to visually check a huge number of surface weather charts. To overcome this problem, we propose an automatic classification system using a support vector machine (SVM), which is a machine-learning method. We attempted to classify pressure patterns into two classes: "winter type" and "non-winter type". For both training datasets and test datasets, we used the JRA-25 dataset from 1981 to 2000. An experimental evaluation showed that our method obtained a greater than 0.8 F-measure. We noted that variations in results were based on differences in training datasets.

KW - Classification

KW - Machine learning

KW - Pressure pattern

KW - Support vector machine (SVM)

UR - http://www.scopus.com/inward/record.url?scp=76449112185&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=76449112185&partnerID=8YFLogxK

U2 - 10.2481/dsj.8.S59

DO - 10.2481/dsj.8.S59

M3 - Article

VL - 8

JO - Data Science Journal

JF - Data Science Journal

SN - 1683-1470

ER -