• Dr. Avinash G M
• Arpitha R
• Chitra Jagadish Badegondra
• Meghana A B
• Pravalika

Agriculture requires effective decisions related to soil, crops, irrigation, fertilizers, diseases, yield, weather, and market conditions. This paper presents an AI-driven Smart Agriculture System that integrates machine learning, IoT, weather services, and web technologies into a unified platform. The system provides soil analysis, crop recommendation, fertilizer recommendation, irrigation prediction, crop yield prediction, plant disease detection, weather monitoring, and market-price analysis. Random Forest, Linear Regression, and MobileNetV3-Small models are used for different agricultural tasks. The system is implemented using React, FastAPI, Python, SQLite, and ESP32-S3-based sensing. Experimental results achieved approximately 92.34% crop accuracy, 95.79% soil accuracy, R² ≈ 0.642 for irrigation, R² ≈ 0.971 for yield prediction, and 91.32% validation accuracy for disease detection. The proposed system demonstrates the integration of AI and IoT for data-driven agricultural decision support.

Artificial Intelligence, Smart Agriculture, Machine Learning, Random Forest, IoT, Soil Analysis, Crop Recommendation, Irrigation, Plant Disease Detection.

"A Smart Decision-Support System for Soil Analysis, Irrigation Scheduling, and Crop-Specific Fertilizer Recommendation Using AI", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.13, Issue 10, page no.b560-b571, October-2026, Available :http://www.jetir.org/papers/JETIR2610158.pdf

"A Smart Decision-Support System for Soil Analysis, Irrigation Scheduling, and Crop-Specific Fertilizer Recommendation Using AI", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.13, Issue 10, page no. ppb560-b571, October-2026, Available at : http://www.jetir.org/papers/JETIR2610158.pdf