EVALUATING FEATURE SELECTION METHODS AND CLASSIFIERS FOR CROP PREDICTION

Read full text Summarise with VIRAT
Authors: Leadership and Sciences International Journal of Technology
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This research evaluates feature selection and machine learning classifiers for crop yield prediction using agricultural datasets. Many feature selection methods were employed to identify which soil, climatic, and agronomic characteristics affected crop yields most. Smaller feature sets reduced redundancy, improved model interpretability, and reduced computational cost. With the chosen features, numerous classifiers were trained and tested. Logistic Regression, Decision Tree, Random Forest. Traditional model performance measurements include prediction accuracy, recall, F1-score, and ROC-AUC. Experimental results suggest that ensemble-based classifiers predict better than solo models. Random Forest exceeded feature subsets in accuracy and robustness. The Decision Tree approach worked well and was simple to use. Using feature selection improves classification outcomes over using all characteristics.

Fields of Study & Subjects

Smart Agriculture and AI Artificial Intelligence in Healthcare Data Mining and Machine Learning Applications

Publication details

Journal / source
IJTLS
Publication date
26 March 2026
Language
EN