Tomato Crop Forecasting: A Comparative Analysis of Regression Models

Tomatoes, renowned for their extensive cultivation and economic significance, hold a pivotal position in agricultural production. This research conducts a comparative analysis of various regression models, namely RF, K-Nearest Neighbors, Support Vector Regression, Lasso Regression, and Linear Regression, evaluating their efficacy in forecasting tomato yields across diverse datasets. The assessment incorporates pivotal metrics such as Mean Absolute Error, Mean Squared Error, Root Mean Squared Error, and R2, alongside computational efficiency metrics like Fit Time (in seconds). Dataset 3 emerges as the most comprehensive and precise dataset compared to Datasets 1 and 2. Despite RF's longer computational time, it demonstrates superior predictive accuracy. Conversely, KNN and Lasso's Regression exhibit commendable performance with lower computational demands. This study aims to support informed decision-making in tomato cultivation, contributing to sustainable agriculture and refined crop yield predictions.