Horse Mounted IoT System for Precision Agriculture and Resource Optimization

Precision agriculture is revolutionizing modern farming by leveraging intelligent decision-making to optimize resource utilization, enhance crop yields, and promote sustainable agricultural practices. This study investigates the application of Support Vector Machine (SVM) and Artificial Neural Network (ANN) classifiers for providing recommendations on water and fertilizer levels based on environmental parameters such as soil moisture, nutrient levels, temperature, humidity, and crop type. A synthetic dataset is developed to train and evaluate the models, ensuring a diverse and representative input for real-world agricultural scenarios. The study compares the classification accuracy, computational efficiency, and adaptability of SVM and ANN in predicting optimal resource requirements for crops. The experimental results indicate that both models offer reliable predictions, with ANN demonstrating superior learning capabilities for complex, nonlinear patterns, while SVM provides robust performance with reduced computational overhead. The findings highlight the advantages and limitations of each classifier, offering insights into their suitability for precision agriculture applications. Additionally, the study emphasizes the role of machine learning in enhancing decision-making processes, reducing resource wastage, and improving agricultural productivity. Future research will focus on integrating deep learning techniques and real-time sensor data to further enhance the predictive accuracy and efficiency of smart farming systems. 

Keywords—Precision Agriculture, Machine Learning, Support Vector Machine (SVM), Artificial Neural Network (ANN), Soil Moisture, Nutrient Levels, Smart Farming, Resource Optimization.