Abstract :

Accurate Crop Prediction Plays A Vital Role In Enhancing The Agricultural Productivity And Enabling Data-driven Decision-making In Smart Farming Systems. Increasing Variability In Soil Characteristics And Environmental Conditions Has Made Crop Selection More Complex, Necessitating Robust And Intelligent Predictive Models. This Paper Proposes A Machine Learning-based Framework For Crop Prediction Using Key Soil Attributes Such As PH, Nitrogen (N), Phosphorus (P), And Potassium (K), Along With Environmental Parameters Including Rainfall And Temperature. A Comprehensive Data Preprocessing Phase Is Employed To Address Class Imbalance Using Synthetic Minority Oversampling Technique (SMOTE) And Majority Weighted Minority Oversampling Technique (MWMOTE). Feature Optimization Is Carried Out Using Boruta, Recursive Feature Elimination (RFE), And A Modified RFE Approach To Retain The Most Informative Features While Reducing Model Complexity. Machine Learning Algorithms Naive Bayes (NB), Decision Tree (DT), K-Nearest Neighbors (KNN), Support Vector Machine (SVM), And Random Forest (RF) Are Evaluated To Assess Predictive Performance. In Addition, Swarm Intelligence Based Optimization Techniques Are Used For Hyperparameter Tuning The Parameters To Further Enhance Model Accuracy. The Proposed Framework Is Extended To Incorporate Market Price Prediction For The Recommended Crop, Offering Practical Insights To Farmers And Agricultural Stakeholders. Experimental Results Demonstrate That Optimized Feature Selection Combined With Ensemble Based Classifiers Significantly Outperforms Conventional Approaches In Terms Of Prediction Accuracy And Robustness.

Published:

09-4-1-2026

Issue:

Vol. 26 No. 4-1 (2026)

Page Nos:

243-249

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite