Unleashing the potential of underutilized datasets to improve agricultural decision-making through comprehensive data analysis: An example of rice crop manager (RCM) dataset

The world is facing a number of challenges such as climate change, diminishing soil quality and stagnant crop yields, which call for scalable solutions to ensure food security for a growing population. In this paper, we explore the role of data science in modern agriculture, highlighting the importa...

Descripción completa

Detalles Bibliográficos
Autores principales: Gakhar, Shalini, Sharma, Sheetal
Formato: Artículo preliminar
Lenguaje:Inglés
Publicado: CGIAR Initiative on Digital Innovation 2023
Materias:
Acceso en línea:https://hdl.handle.net/10568/139411
_version_ 1855530330668138496
author Gakhar, Shalini
Sharma, Sheetal
author_browse Gakhar, Shalini
Sharma, Sheetal
author_facet Gakhar, Shalini
Sharma, Sheetal
author_sort Gakhar, Shalini
collection Repository of Agricultural Research Outputs (CGSpace)
description The world is facing a number of challenges such as climate change, diminishing soil quality and stagnant crop yields, which call for scalable solutions to ensure food security for a growing population. In this paper, we explore the role of data science in modern agriculture, highlighting the importance of big data analytics, geospatial technology, and machine learning. We use the datasets collected in a decision support tool called Rice Crop Manager (RCM) and employ machine learning algorithms to estimate yield targets, which can then be used to generate field-specific nutrient management recommendations. The proposed methodology involves extracting useful insights from the data by using statistics, artificial intelligence (AI), domain knowledge, and predictive analysis. We use cutting-edge technologies like machine learning to develop a model that can select the optimal parameters to generate the recommendations. We estimate the target yield with a calibration and validation ratio of 70:30 through performance evaluation, particularly by computing R-Square, root mean square error, and mean absolute error. Our analysis shows that this approach is well-suited for the acquired data, and it could be automated to derive more valuable information for the benefit of farmers. We propose expanding RCM's functionality to include machine learning for estimating target yields, with the aim of optimizing nutrient management recommendations and reducing decision-making time for farmers and stakeholders. Our study emphasizes the untapped potential of underutilized spatial and temporal data collected by decision-making tools, positioning machine learning as a catalyst for improved resource management in agriculture. The findings contribute to ongoing efforts in sustainable and innovative agricultural practices.
format Artículo preliminar
id CGSpace139411
institution CGIAR Consortium
language Inglés
publishDate 2023
publishDateRange 2023
publishDateSort 2023
publisher CGIAR Initiative on Digital Innovation
publisherStr CGIAR Initiative on Digital Innovation
record_format dspace
spelling CGSpace1394112024-11-07T09:39:00Z Unleashing the potential of underutilized datasets to improve agricultural decision-making through comprehensive data analysis: An example of rice crop manager (RCM) dataset Gakhar, Shalini Sharma, Sheetal data data collection machine learning artificial intelligence yield forecasting data analysis The world is facing a number of challenges such as climate change, diminishing soil quality and stagnant crop yields, which call for scalable solutions to ensure food security for a growing population. In this paper, we explore the role of data science in modern agriculture, highlighting the importance of big data analytics, geospatial technology, and machine learning. We use the datasets collected in a decision support tool called Rice Crop Manager (RCM) and employ machine learning algorithms to estimate yield targets, which can then be used to generate field-specific nutrient management recommendations. The proposed methodology involves extracting useful insights from the data by using statistics, artificial intelligence (AI), domain knowledge, and predictive analysis. We use cutting-edge technologies like machine learning to develop a model that can select the optimal parameters to generate the recommendations. We estimate the target yield with a calibration and validation ratio of 70:30 through performance evaluation, particularly by computing R-Square, root mean square error, and mean absolute error. Our analysis shows that this approach is well-suited for the acquired data, and it could be automated to derive more valuable information for the benefit of farmers. We propose expanding RCM's functionality to include machine learning for estimating target yields, with the aim of optimizing nutrient management recommendations and reducing decision-making time for farmers and stakeholders. Our study emphasizes the untapped potential of underutilized spatial and temporal data collected by decision-making tools, positioning machine learning as a catalyst for improved resource management in agriculture. The findings contribute to ongoing efforts in sustainable and innovative agricultural practices. 2023-12-07 2024-02-15T05:47:02Z 2024-02-15T05:47:02Z Working Paper https://hdl.handle.net/10568/139411 en Open Access application/pdf CGIAR Initiative on Digital Innovation Gakhar, Shalini, Sheetal Sharma. 2023. Unleashing the potential of underutilized datasets to improve agricultural decision-making through comprehensive data analysis: An example of rice crop manager (RCM) dataset. Working Paper 5. Enabling digital platforms & services, CGIAR Initiative on Digital Innovation. 20 p.
spellingShingle data
data collection
machine learning
artificial intelligence
yield forecasting
data analysis
Gakhar, Shalini
Sharma, Sheetal
Unleashing the potential of underutilized datasets to improve agricultural decision-making through comprehensive data analysis: An example of rice crop manager (RCM) dataset
title Unleashing the potential of underutilized datasets to improve agricultural decision-making through comprehensive data analysis: An example of rice crop manager (RCM) dataset
title_full Unleashing the potential of underutilized datasets to improve agricultural decision-making through comprehensive data analysis: An example of rice crop manager (RCM) dataset
title_fullStr Unleashing the potential of underutilized datasets to improve agricultural decision-making through comprehensive data analysis: An example of rice crop manager (RCM) dataset
title_full_unstemmed Unleashing the potential of underutilized datasets to improve agricultural decision-making through comprehensive data analysis: An example of rice crop manager (RCM) dataset
title_short Unleashing the potential of underutilized datasets to improve agricultural decision-making through comprehensive data analysis: An example of rice crop manager (RCM) dataset
title_sort unleashing the potential of underutilized datasets to improve agricultural decision making through comprehensive data analysis an example of rice crop manager rcm dataset
topic data
data collection
machine learning
artificial intelligence
yield forecasting
data analysis
url https://hdl.handle.net/10568/139411
work_keys_str_mv AT gakharshalini unleashingthepotentialofunderutilizeddatasetstoimproveagriculturaldecisionmakingthroughcomprehensivedataanalysisanexampleofricecropmanagerrcmdataset
AT sharmasheetal unleashingthepotentialofunderutilizeddatasetstoimproveagriculturaldecisionmakingthroughcomprehensivedataanalysisanexampleofricecropmanagerrcmdataset