American
International
University-
Bangladesh

Students & Supervisors

Abstract

In accurate agriculture by enabling data-driven decision-making to enhance crop production and investment management artificial intelligence is a role of pattern. This study supports the "Smart Agricultural Production Optimizing Engine" dataset, comprising critical agronomic features such as nitrogen, phosphorus, potassium levels, soil pH, temperature, humidity, and rainfall, to develop machine learning models for crop appropriate forecast. The system information for agricultural site was carefully process through normalization, missing value imputation, and feature scaling to improve model accuracy. Several algorithms were judge, including Naive Bayes, Random Forest, Decision Tree, and Logistic Regression. Among these, the Naive Bayes sorter indicate the highest prediction accuracy of 92%, with precision and recollect values of 90% and 89%, accordingly, validating its consistent performance in relate optimal crop types based on environmental load. Random Forest and Decision Tree models also showed rival outcomes with accuracies around 85% and 83%, and F1-scores of 0.84 and 0.81, individually. Logistic Regression achieved moderate accuracy at approximately 78%. The models smooth variable rate planting and fertilization direct customize to specific field conditions, potentially enhancing crop yield by over 15% while reducing fertilizer wastage by around 12%. moreover, the relation of AI with IoT sensor data enabled real-time irrigated timeline, improving water-use planning by an estimated 20%. Enhanced monitoring and targeted interventions contributed to a projected 10% reduction in nutrient runoff and soil degradation, supporting environmentally sustainable farming practices. This complete data-driven approach amplifies resource use, reduce environmental impact, and allow farmers with accuracy, practical findings. AI-driven precision farming, supported by real- world datasets and machine learning, offers exceptional items in improving agricultural development and sustainability. Future work plan for making the agricultural sites are personalized learning techniques to further improve model correctness and strength across diverse agricultural environments."

Keywords

Publication Details

• Type of Publication:
• Conference Name: Gazipur Agricultural University International Conference (GAUIC 2025)
• Date of Conference: 12/12/2025 - 12/12/2025
• Venue: Gazipur 1706, Bangladesh
• Organizer: Gazipur Agricultural University,Gazipur,Bangladesh