ADASYN-LOF Algorithm for Imbalanced Tornado Samples
Early warning and forecasting of tornadoes began to combine artificial intelligence (AI) and machine learning (ML) algorithms to improve identification efficiency in the past few years. Applying machine learning algorithms to detect tornadoes usually encounters class imbalance problems because torna...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-03-01
|
| Series: | Atmosphere |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4433/13/4/544 |
| _version_ | 1827621760267190272 |
|---|---|
| author | Zhipeng Qing Qiangyu Zeng Hao Wang Yin Liu Taisong Xiong Shihao Zhang |
| author_facet | Zhipeng Qing Qiangyu Zeng Hao Wang Yin Liu Taisong Xiong Shihao Zhang |
| author_sort | Zhipeng Qing |
| collection | DOAJ |
| description | Early warning and forecasting of tornadoes began to combine artificial intelligence (AI) and machine learning (ML) algorithms to improve identification efficiency in the past few years. Applying machine learning algorithms to detect tornadoes usually encounters class imbalance problems because tornadoes are rare events in weather processes. The ADASYN-LOF algorithm (ALA) was proposed to solve the imbalance problem of tornado sample sets based on radar data. The adaptive synthetic (ADASYN) sampling algorithm is used to solve the imbalance problem by increasing the number of minority class samples, combined with the local outlier factor (LOF) algorithm to denoise the synthetic samples. The performance of the ALA algorithm is tested by using the supporting vector machine (SVM), artificial neural network (ANN), and random forest (RF) models. The results show that the ALA algorithm can improve the performance and noise immunity of the models, significantly increase the tornado recognition rate, and have the potential to increase the early tornado warning time. ALA is more effective in preprocessing imbalanced data of SVM and ANN, compared with ADASYN, Synthetic Minority Oversampling Technique (SMOTE), SMOTE-LOF algorithms. |
| first_indexed | 2024-03-09T11:09:51Z |
| format | Article |
| id | doaj.art-30c31e4bcc4e428886e745e9317be2ac |
| institution | Directory Open Access Journal |
| issn | 2073-4433 |
| language | English |
| last_indexed | 2024-03-09T11:09:51Z |
| publishDate | 2022-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Atmosphere |
| spelling | doaj.art-30c31e4bcc4e428886e745e9317be2ac2023-12-01T00:46:22ZengMDPI AGAtmosphere2073-44332022-03-0113454410.3390/atmos13040544ADASYN-LOF Algorithm for Imbalanced Tornado SamplesZhipeng Qing0Qiangyu Zeng1Hao Wang2Yin Liu3Taisong Xiong4Shihao Zhang5CMA Key Laboratory of Atmospheric Sounding, Chengdu 610225, ChinaCMA Key Laboratory of Atmospheric Sounding, Chengdu 610225, ChinaCMA Key Laboratory of Atmospheric Sounding, Chengdu 610225, ChinaJiangsu Meteorological Observation Center, Nanjing 210041, ChinaCMA Key Laboratory of Atmospheric Sounding, Chengdu 610225, ChinaCMA Key Laboratory of Atmospheric Sounding, Chengdu 610225, ChinaEarly warning and forecasting of tornadoes began to combine artificial intelligence (AI) and machine learning (ML) algorithms to improve identification efficiency in the past few years. Applying machine learning algorithms to detect tornadoes usually encounters class imbalance problems because tornadoes are rare events in weather processes. The ADASYN-LOF algorithm (ALA) was proposed to solve the imbalance problem of tornado sample sets based on radar data. The adaptive synthetic (ADASYN) sampling algorithm is used to solve the imbalance problem by increasing the number of minority class samples, combined with the local outlier factor (LOF) algorithm to denoise the synthetic samples. The performance of the ALA algorithm is tested by using the supporting vector machine (SVM), artificial neural network (ANN), and random forest (RF) models. The results show that the ALA algorithm can improve the performance and noise immunity of the models, significantly increase the tornado recognition rate, and have the potential to increase the early tornado warning time. ALA is more effective in preprocessing imbalanced data of SVM and ANN, compared with ADASYN, Synthetic Minority Oversampling Technique (SMOTE), SMOTE-LOF algorithms.https://www.mdpi.com/2073-4433/13/4/544tornadoesclass imbalancemachine learning |
| spellingShingle | Zhipeng Qing Qiangyu Zeng Hao Wang Yin Liu Taisong Xiong Shihao Zhang ADASYN-LOF Algorithm for Imbalanced Tornado Samples Atmosphere tornadoes class imbalance machine learning |
| title | ADASYN-LOF Algorithm for Imbalanced Tornado Samples |
| title_full | ADASYN-LOF Algorithm for Imbalanced Tornado Samples |
| title_fullStr | ADASYN-LOF Algorithm for Imbalanced Tornado Samples |
| title_full_unstemmed | ADASYN-LOF Algorithm for Imbalanced Tornado Samples |
| title_short | ADASYN-LOF Algorithm for Imbalanced Tornado Samples |
| title_sort | adasyn lof algorithm for imbalanced tornado samples |
| topic | tornadoes class imbalance machine learning |
| url | https://www.mdpi.com/2073-4433/13/4/544 |
| work_keys_str_mv | AT zhipengqing adasynlofalgorithmforimbalancedtornadosamples AT qiangyuzeng adasynlofalgorithmforimbalancedtornadosamples AT haowang adasynlofalgorithmforimbalancedtornadosamples AT yinliu adasynlofalgorithmforimbalancedtornadosamples AT taisongxiong adasynlofalgorithmforimbalancedtornadosamples AT shihaozhang adasynlofalgorithmforimbalancedtornadosamples |