Predicting crop phenology: a simple logistic regression model approach

Crop yield prediction plays a central role in the agricultural planning and decision-making processes. In this paper, we analyze the phenology as a crucial aspect of this topic. We propose a simple model to predict phenology groups on maize and wheat crops at the field-level in Argentina. Our model...

Descripción completa

Detalles Bibliográficos
Autores principales: Leale, Guillermo, Cocitto, Bruno, Cardoso, Ana Laura, Lafluf, Pedro, Tantucci, Ligia, Mendez, Fernanda
Formato: Objeto de conferencia
Lenguaje:Inglés
Publicado: 2023
Materias:
Ciencias Informáticas
phenology prediction
logistic regression
photoperiod
Acceso en línea:http://sedici.unlp.edu.ar/handle/10915/165462
Aporte de:
SEDICI (UNLP) de Universidad Nacional de La Plata
id I19-R120-10915-165462
record_format dspace
spelling I19-R120-10915-1654622024-04-29T20:05:14Z http://sedici.unlp.edu.ar/handle/10915/165462 Predicting crop phenology: a simple logistic regression model approach Leale, Guillermo Cocitto, Bruno Cardoso, Ana Laura Lafluf, Pedro Tantucci, Ligia Mendez, Fernanda 2023-09 2023 2024-04-29T16:56:53Z en Ciencias Informáticas phenology prediction logistic regression photoperiod Crop yield prediction plays a central role in the agricultural planning and decision-making processes. In this paper, we analyze the phenology as a crucial aspect of this topic. We propose a simple model to predict phenology groups on maize and wheat crops at the field-level in Argentina. Our model uses logistic regression and includes photoperiod as an explanatory variable, which is very simple to calculate taking into account latitude and date as input. A large number of data records are used to obtain accurate results. Our model has been tested with over 77% accuracy for both crops. It was also benchmarked with Random Forest, which gives comparable results. However, our study shows that a very simple approach could be used with logistic regression, with very little loss of performance. Our model obtains phenology groups and also performs well with certain critical phenology stages for both crops. Our study aims to provide a simple and effective method for predicting phenology, which can be an aid to crop prediction and for farmers to make accurate decisions. Our work emphasizes the simplicity of the model, the use of a large number of data records, and the inclusion of the photoperiod as an input variable. Sociedad Argentina de Informática e Investigación Operativa Objeto de conferencia Objeto de conferencia http://creativecommons.org/licenses/by-nc-sa/4.0/ Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) application/pdf 111-124
institution Universidad Nacional de La Plata
institution_str I-19
repository_str R-120
collection SEDICI (UNLP)
language Inglés
topic Ciencias Informáticas
phenology prediction
logistic regression
photoperiod
spellingShingle Ciencias Informáticas
phenology prediction
logistic regression
photoperiod
Leale, Guillermo
Cocitto, Bruno
Cardoso, Ana Laura
Lafluf, Pedro
Tantucci, Ligia
Mendez, Fernanda
Predicting crop phenology: a simple logistic regression model approach
topic_facet Ciencias Informáticas
phenology prediction
logistic regression
photoperiod
description Crop yield prediction plays a central role in the agricultural planning and decision-making processes. In this paper, we analyze the phenology as a crucial aspect of this topic. We propose a simple model to predict phenology groups on maize and wheat crops at the field-level in Argentina. Our model uses logistic regression and includes photoperiod as an explanatory variable, which is very simple to calculate taking into account latitude and date as input. A large number of data records are used to obtain accurate results. Our model has been tested with over 77% accuracy for both crops. It was also benchmarked with Random Forest, which gives comparable results. However, our study shows that a very simple approach could be used with logistic regression, with very little loss of performance. Our model obtains phenology groups and also performs well with certain critical phenology stages for both crops. Our study aims to provide a simple and effective method for predicting phenology, which can be an aid to crop prediction and for farmers to make accurate decisions. Our work emphasizes the simplicity of the model, the use of a large number of data records, and the inclusion of the photoperiod as an input variable.
format Objeto de conferencia
Objeto de conferencia
author Leale, Guillermo
Cocitto, Bruno
Cardoso, Ana Laura
Lafluf, Pedro
Tantucci, Ligia
Mendez, Fernanda
author_facet Leale, Guillermo
Cocitto, Bruno
Cardoso, Ana Laura
Lafluf, Pedro
Tantucci, Ligia
Mendez, Fernanda
author_sort Leale, Guillermo
title Predicting crop phenology: a simple logistic regression model approach
title_short Predicting crop phenology: a simple logistic regression model approach
title_full Predicting crop phenology: a simple logistic regression model approach
title_fullStr Predicting crop phenology: a simple logistic regression model approach
title_full_unstemmed Predicting crop phenology: a simple logistic regression model approach
title_sort predicting crop phenology: a simple logistic regression model approach
publishDate 2023
url http://sedici.unlp.edu.ar/handle/10915/165462
work_keys_str_mv AT lealeguillermo predictingcropphenologyasimplelogisticregressionmodelapproach
AT cocittobruno predictingcropphenologyasimplelogisticregressionmodelapproach
AT cardosoanalaura predictingcropphenologyasimplelogisticregressionmodelapproach
AT laflufpedro predictingcropphenologyasimplelogisticregressionmodelapproach
AT tantucciligia predictingcropphenologyasimplelogisticregressionmodelapproach
AT mendezfernanda predictingcropphenologyasimplelogisticregressionmodelapproach
_version_ 1807223057824088064

Ejemplares similares