Crop Prediction & Production System Using ML Approaches
Prof. Shwetha L
Asst Professor, Dept. of CSE
G. Madegowda Institute of Technology
Bharathinagara, Mandya – 571 422
India
shwethal.gmitcse@gmail.com
Dr. Nuthan A C
Professor & Head, Dept. of AI&ML
G. Madegowda Institute of Technology
Bharathinagara, Mandya – 571 422
India
aimlhod.gmit@gmail.com
Ankitha N
Student, Dept. of AI&ML
G. Madegowda Institute of Technology
Bharathinagara, Mandya – 571 422
India
ankithananjappaankitha@gmail.com
Priya A C
Student, Dept. of AI&ML
G. Madegowda Institute of Technology
Bharathinagara, Mandya – 571 422
India
lopagowda22@gmail.com
Sinchana N S
Student, Dept. of AI&ML
G. Madegowda Institute of Technology
Bharathinagara, Mandya – 571 422
India
sinchanashivanaga@gmail.com
Abstract - This Crop Prediction and Production System that leverages Machine Learning to provide data-driven agricultural decision support tailored for Indian farming communities. The system employs a Random Forest classifier, achieving 87.5% accuracy by analyzing seven critical parameters: soil nutrients (N, P, K), temperature, humidity, pH, and rainfall to recommend optimal crops with confidence scoring. A key innovation is its multi-language interface supporting five Indian languages (English, Hindi, Kannada, Telugu, Tamil), effectively bridging the digital divide for non-English speaking farmers. The platform uniquely integrates predictive analytics with a digital marketplace for real-time price intelligence, land analysis tools for soil health and irrigation planning, and region-specific advisory services. Designed for accessibility, it features QR code-based mobile access, responsive design for low-bandwidth environments, and batch processing capabilities.
Implemented using Flask framework with Bootstrap frontend, the system demonstrates practical deployment of ML in agriculture, enhancing productivity through scientific crop selection while maintaining cultural and technological relevance for diverse user groups. Testing with 50 farmers showed 82% satisfaction with predictions and strong preference for regional language interfaces.
Key Words: Machine Learning, Agricultural Technology, Crop Prediction, Random Forest, Multi-language Interface, Precision Agriculture, Digital Marketplace, Soil Analysis, Mobile Accessibility, Indian Farming, Flask Framework, Decision Support System, Agricultural Informatics, Sustainable Farming, QR Code Integration.
