XGB+FM for Severe Convection Forecast and Factor Selection
In the field of meteorology, radiosonde data and observation data are critical for analyzing regional meteorological characteristics. Because of the high false alarm rate, severe convection forecasting is still challenging. In addition, the existing methods are difficult to use to capture the intera...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-01-01
|
Series: | Electronics |
Subjects: | |
Online Access: | https://www.mdpi.com/2079-9292/10/3/321 |
_version_ | 1797405855424970752 |
---|---|
author | Zhiying Lu Xudong Ding Xin Li Haopeng Wu Xiaolei Sun |
author_facet | Zhiying Lu Xudong Ding Xin Li Haopeng Wu Xiaolei Sun |
author_sort | Zhiying Lu |
collection | DOAJ |
description | In the field of meteorology, radiosonde data and observation data are critical for analyzing regional meteorological characteristics. Because of the high false alarm rate, severe convection forecasting is still challenging. In addition, the existing methods are difficult to use to capture the interaction of meteorological factors at the same time. In this research, a cascade of extreme gradient boosting (XGBoost) for feature transformation and a factorization machine (FM) for second-order feature interaction to capture the nonlinear interaction—XGB+FM—is proposed. An attention-based bidirectional long short-term memory (Att-Bi-LSTM) network is proposed to impute the missing data of meteorological observation stations. The problem of class imbalance is resolved by the support vector machines–synthetic minority oversampling technique (SVM-SMOTE), in which two oversampling strategies based on the support vector discrimination mechanism are proposed. It is proven that the method is effective, and the threat score (TS) is 7.27~14.28% higher than other methods. Moreover, we propose the meteorological factor selection method based on XGB+FM and improve the forecast accuracy, which is one of our contributions, as well as the forecast system. |
first_indexed | 2024-03-09T03:17:16Z |
format | Article |
id | doaj.art-bdafc9bc52494245bf1c8dc7f301e5d4 |
institution | Directory Open Access Journal |
issn | 2079-9292 |
language | English |
last_indexed | 2024-03-09T03:17:16Z |
publishDate | 2021-01-01 |
publisher | MDPI AG |
record_format | Article |
series | Electronics |
spelling | doaj.art-bdafc9bc52494245bf1c8dc7f301e5d42023-12-03T15:18:06ZengMDPI AGElectronics2079-92922021-01-0110332110.3390/electronics10030321XGB+FM for Severe Convection Forecast and Factor SelectionZhiying Lu0Xudong Ding1Xin Li2Haopeng Wu3Xiaolei Sun4School of Electrical and Information Engineering, Tianjin University, Tianjin 300072, ChinaSchool of Electrical and Information Engineering, Tianjin University, Tianjin 300072, ChinaSchool of Electrical and Information Engineering, Tianjin University, Tianjin 300072, ChinaSchool of Electrical and Information Engineering, Tianjin University, Tianjin 300072, ChinaTianjin Bureau of Meteorology, Tianjin 300074, ChinaIn the field of meteorology, radiosonde data and observation data are critical for analyzing regional meteorological characteristics. Because of the high false alarm rate, severe convection forecasting is still challenging. In addition, the existing methods are difficult to use to capture the interaction of meteorological factors at the same time. In this research, a cascade of extreme gradient boosting (XGBoost) for feature transformation and a factorization machine (FM) for second-order feature interaction to capture the nonlinear interaction—XGB+FM—is proposed. An attention-based bidirectional long short-term memory (Att-Bi-LSTM) network is proposed to impute the missing data of meteorological observation stations. The problem of class imbalance is resolved by the support vector machines–synthetic minority oversampling technique (SVM-SMOTE), in which two oversampling strategies based on the support vector discrimination mechanism are proposed. It is proven that the method is effective, and the threat score (TS) is 7.27~14.28% higher than other methods. Moreover, we propose the meteorological factor selection method based on XGB+FM and improve the forecast accuracy, which is one of our contributions, as well as the forecast system.https://www.mdpi.com/2079-9292/10/3/321severe convection forecastXGBoostFMAtt-Bi-LSTMSVM-SMOTEBayesian optimization |
spellingShingle | Zhiying Lu Xudong Ding Xin Li Haopeng Wu Xiaolei Sun XGB+FM for Severe Convection Forecast and Factor Selection Electronics severe convection forecast XGBoost FM Att-Bi-LSTM SVM-SMOTE Bayesian optimization |
title | XGB+FM for Severe Convection Forecast and Factor Selection |
title_full | XGB+FM for Severe Convection Forecast and Factor Selection |
title_fullStr | XGB+FM for Severe Convection Forecast and Factor Selection |
title_full_unstemmed | XGB+FM for Severe Convection Forecast and Factor Selection |
title_short | XGB+FM for Severe Convection Forecast and Factor Selection |
title_sort | xgb fm for severe convection forecast and factor selection |
topic | severe convection forecast XGBoost FM Att-Bi-LSTM SVM-SMOTE Bayesian optimization |
url | https://www.mdpi.com/2079-9292/10/3/321 |
work_keys_str_mv | AT zhiyinglu xgbfmforsevereconvectionforecastandfactorselection AT xudongding xgbfmforsevereconvectionforecastandfactorselection AT xinli xgbfmforsevereconvectionforecastandfactorselection AT haopengwu xgbfmforsevereconvectionforecastandfactorselection AT xiaoleisun xgbfmforsevereconvectionforecastandfactorselection |