Smart Soil-Based Crop Recommendation System Using Real-Time Soil and Climate Data with Machine Learning

This paper has presented a Smart Soil-Based Crop Recommendation System that successfully integrates GPS-based location detection, government soil API integration, open-access climate data retrieval, Random Forest machine learning prediction, fertilizer planning, and Tamil voice output into a single, sensor-free, web-based agricultural decision support platform. The system achieves 93.2% classification accuracy on a 56-class crop recommendation task, delivers end-to-end recommendations in under 3 seconds on broadband connections, covers all 38 districts of Tamil Nadu, and requires no physical hardware beyond a GPS-enabled device. The work demonstrates that real-time, location-specific crop recommendations of practical utility can be delivered to farmers at zero hardware cost by leveraging freely available government and open-source APIs. The traditional crop boost mechanism successfully integrates regional agricultural knowledge with data-driven prediction, and Tamil voice output extends system accessibility to users with limited digital literacy. These results confirm the viability of API-driven, ML-powered agricultural decision support as a scalable and inclusive approach to precision farming in resource-constrained environments.

[1] M. D. Shaifullah Sharafat et al., "Real-Time Crop Prediction Using Sensor Data and Weather APIs with Edge Computing," International Journal of Agricultural Technology, vol. 19, no. 3, pp. 1021–1038, 2023. [2] B. Dey et al., "Crop Recommendation Using Supervised Machine Learning with Soil and Climate Data," International Journal of Computer Applications, vol. 184, no. 12, pp. 1–8, 2022. [3] S. Shafiulla et al., "Crop Recommendation using Machine Learning Techniques," International Journal of Engineering Research & Technology (IJERT), vol. 10, no. 11, pp. 1–6, 2022. [4] S. Bhargavi and Dr. Srinivasan, "Crop Recommendation System Using Machine Learning," IJERT, vol. 11, no. 6, pp. 1–5, 2023. [5] K. Jaya Deepthi et al., "Smart Agriculture: Crop Recommendation and Yield Prediction Using Random Forest," in Proc. Int. Conf. on Intelligent Transportation Systems and Mobility Management (ICITSM), pp. 1–8, 2025. [6] R. Karthikeyan et al., "Implementation of Effective Crop Selection Using Random Forest Algorithm," International Journal of Engineering and Technology (IJET), vol. 7, no. 3, pp. 34–39, 2018. [7] S. B. Bukhari, "Crop Recommendation System: A Data-Driven Approach," Journal of Techno Trainers, vol. 1, no. 2, pp. 26–35, 2024. [8] B. N. S. Gupta and U. Lekhana, "Crop Recommender System Using ML Approach," International Journal of Emerging Research in Science and Technology (IJERST), vol. 20, no. 2, pp. 1167–1177, 2024. [9] R. S. Nair et al., "Smart Crop Selection: ML-Based Decision Support System," International Journal for Research in Applied Science and Engineering Technology (IJRASET), pp. 1–6, 2025. [10] H. Afzal et al., "Soil Information with Machine Learning for Crop Recommendation," Scientific Reports, vol. 15, Article 8560, 2025. doi: 10.1038/s41598-025-92345-6.