Machine Learning with Voting Committee for Frost Prediction
A machine learning (ML)-based methodology for predicting frosts was applied to the southern and southeastern regions of Brazil, as well as to other countries including Uruguay, Paraguay, northern Argentina, and southeastern Bolivia. The machine learning model (using TensorFlow (TF)) was compared to...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-02-01
|
| Series: | Meteorology |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2674-0494/4/1/6 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850279658808934400 |
|---|---|
| author | Vinícius Albuquerque de Almeida Juliana Aparecida Anochi José Roberto Rozante Haroldo Fraga de Campos Velho |
| author_facet | Vinícius Albuquerque de Almeida Juliana Aparecida Anochi José Roberto Rozante Haroldo Fraga de Campos Velho |
| author_sort | Vinícius Albuquerque de Almeida |
| collection | DOAJ |
| description | A machine learning (ML)-based methodology for predicting frosts was applied to the southern and southeastern regions of Brazil, as well as to other countries including Uruguay, Paraguay, northern Argentina, and southeastern Bolivia. The machine learning model (using TensorFlow (TF)) was compared to the frost index (IG from the Portuguese: Índice de Geada) developed by the National Institute for Space Research (INPE, Brazil). The IG is estimated using meteorological variables from a regional weather numerical model (RWNM). After calculating the two indices using the ML model and the RWNM, a voting committee (VC) was trained to select between the computed outputs. The AdaBoostClassifier algorithm was employed to implement the voting committee. The study area was subdivided into three distinct subregions: R1 (outside Brazil), R2 (the south of Brazil), and R3 (southeastern Brazil). Two forecasting time scales were evaluated: 24 h and 72 h. The 24 h forecasts from both approaches (TF and RWNM) exhibited a similar performance in terms of the number of accurate predictions. However, in the region covering Uruguay and northern Argentina, the TensorFlow model demonstrated superior frost prediction accuracy. Additionally, the TensorFlow model outperformed the RWNM for the 72 h forecast horizon. |
| format | Article |
| id | doaj-art-8e1a871dfdbe478abdf0be9be62c2677 |
| institution | OA Journals |
| issn | 2674-0494 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Meteorology |
| spelling | doaj-art-8e1a871dfdbe478abdf0be9be62c26772025-08-20T01:49:01ZengMDPI AGMeteorology2674-04942025-02-0141610.3390/meteorology4010006Machine Learning with Voting Committee for Frost PredictionVinícius Albuquerque de Almeida0Juliana Aparecida Anochi1José Roberto Rozante2Haroldo Fraga de Campos Velho3Laboratory for Applied Meteorology, Federal University of Rio de Janeiro, Rio de Janeiro 21941-901, BrazilNational Institute for Space Research, São José dos Campos 12227-010, BrazilNational Institute for Space Research, São José dos Campos 12227-010, BrazilNational Institute for Space Research, São José dos Campos 12227-010, BrazilA machine learning (ML)-based methodology for predicting frosts was applied to the southern and southeastern regions of Brazil, as well as to other countries including Uruguay, Paraguay, northern Argentina, and southeastern Bolivia. The machine learning model (using TensorFlow (TF)) was compared to the frost index (IG from the Portuguese: Índice de Geada) developed by the National Institute for Space Research (INPE, Brazil). The IG is estimated using meteorological variables from a regional weather numerical model (RWNM). After calculating the two indices using the ML model and the RWNM, a voting committee (VC) was trained to select between the computed outputs. The AdaBoostClassifier algorithm was employed to implement the voting committee. The study area was subdivided into three distinct subregions: R1 (outside Brazil), R2 (the south of Brazil), and R3 (southeastern Brazil). Two forecasting time scales were evaluated: 24 h and 72 h. The 24 h forecasts from both approaches (TF and RWNM) exhibited a similar performance in terms of the number of accurate predictions. However, in the region covering Uruguay and northern Argentina, the TensorFlow model demonstrated superior frost prediction accuracy. Additionally, the TensorFlow model outperformed the RWNM for the 72 h forecast horizon.https://www.mdpi.com/2674-0494/4/1/6frost indexfrost predictiondeep learningcommittee machine |
| spellingShingle | Vinícius Albuquerque de Almeida Juliana Aparecida Anochi José Roberto Rozante Haroldo Fraga de Campos Velho Machine Learning with Voting Committee for Frost Prediction Meteorology frost index frost prediction deep learning committee machine |
| title | Machine Learning with Voting Committee for Frost Prediction |
| title_full | Machine Learning with Voting Committee for Frost Prediction |
| title_fullStr | Machine Learning with Voting Committee for Frost Prediction |
| title_full_unstemmed | Machine Learning with Voting Committee for Frost Prediction |
| title_short | Machine Learning with Voting Committee for Frost Prediction |
| title_sort | machine learning with voting committee for frost prediction |
| topic | frost index frost prediction deep learning committee machine |
| url | https://www.mdpi.com/2674-0494/4/1/6 |
| work_keys_str_mv | AT viniciusalbuquerquedealmeida machinelearningwithvotingcommitteeforfrostprediction AT julianaaparecidaanochi machinelearningwithvotingcommitteeforfrostprediction AT joserobertorozante machinelearningwithvotingcommitteeforfrostprediction AT haroldofragadecamposvelho machinelearningwithvotingcommitteeforfrostprediction |