Oluwaseunlabisi/Big-Data-Driven-Crop-Production-Forecasting-for-Improved-Agricultural-Management-in-Nigeria
Domain:
agricultureRecord type:
projectdataset
Creator:
Olu
Host:
# Big Data-Driven Crop Production Forecasting for Improved Agricultural Management in Nigeria
*Overview*
This repository contains the code, datasets, and resources for the project "Big Data-Driven Crop Production Forecasting for Improved Agricultural Management in Nigeria". The project aims to improve agricultural production forecasting accuracy by integrating climate variables such as temperature, precipitation, vapor pressure, and evapotranspiration with traditional crop production data. This approach overcomes the limitations of univariate models, providing a more comprehensive understanding of agricultural trends.
*Contents*
- data: Datasets used in the project, including:
FAOSTAT_data_en_7-31-2024.csv: Historical crop production data.
crucy.v4.07.1901.2022.Nigeria.pet.per: Climate data (evapotranspiration),
crucy.v4.07.1901.2022.Nigeria.pre.per: Climate data (precipitation),
crucy.v4.07.1901.2022.Nigeria.tmp.per: Climate data (temperature),
crucy.v4.07.1901.2022.Nigeria.vap.per: Climate data (vapour pressure).
aggregate_data.csv: A yearly aggregated dataset that contains the combination of historical crop data and climate data.
- notebooks: Jupyter notebooks containing the code for data preprocessing and, analysis & model development.
Data_preprocessing_ProjectCode.ipynb: Code for cleaning and integrating crop production and climate datasets.
model_training.ipynb: Code for training the multivariate forecasting models, including LSTM, CNN, MLP, and XGBoost.
Crop_forecast_ProjectCode_2.ipynb: Analysis of model performance and visualization of forecasting results.
*Key Features*
- MapReduce for Data Aggregation (PySpark)
- Multivariate time series forecasting model that integrates climate data.
- Comparison of advanced machine learning models (LSTM, CNN, MLP) with traditional models like XGBoost.
- Model evaluation and selection based on forecasting accuracy and adaptability to changing climate conditions.
- Forecast of agricultural production for the next …
Similar
snyamson/Data-Driven-Crop-Recommendation-for-Agricultural-AdvancementLIBSENSE-RUFORUM Collaboration for Improved Agricultural Research Data Management in AfricaReplication Data for: "Data-driven decentralized breeding increases genetic gain in challenging crop production environments"Data-Driven Weather Forecasting in South African Farming: Impacts on Crop YieldsEstimating Agricultural and Forest Lands for Improved Post COVID-19 Crop Production Using Open Remote Sensing: Case Study of Kwara State, NigeriaData-driven decentralized breeding increases prediction accuracy in a challenging crop production environment