Year: 2024
Type: article
Abstract: The research-based project aims to develop a predictive model for cardamom production in Sikkim using historical rainfall data and the area under cultivation.Given the critical role of agriculture in the region, this project provides a valuable tool for farmers and agricultural planners to forecast production and make informed decisions.The dataset used for this study was obtained from various agricultural records and includes information on actual rainfall in different periods and cardamom production metrics.A linear regression model was trained to predict cardamom production based on four rainfall periods and the area under cultivation.The model's performance was evaluated using metrics such as Mean Absolute Error (MAE), Mean Squared Error (MSE), and R-squared (R²), yielding promising results with an R² value of 0.80, indicating a strong predictive capability.To enhance user accessibility, a graphical user interface (GUI) was developed; this interface allows users to input rainfall and area data to obtain predictions for cardamom production.It also provides a detailed analysis page where users can view the actual data, standardized data, and correlation matrices.Overall, this project demonstrates a comprehensive approach to data analysis and model development, offering a practical application for predicting agricultural yields.The integration of a user-friendly interface further extends the project's utility, making it a valuable resource for stakeholders in Sikkim's agricultural sector.
Author: Purnima Chettri
Language: English
Topic: Smart Agriculture and AI
Subfield: Plant Science
Domain: Life Sciences
Open Access status: hybrid
OpenAlex ID: W4411515128