Result Paper on Animal Disease Prediction System Using Machine Learning (CNN and SVM)

Abstract

Animal disease prediction is a critical challenge in veterinary science and agricultural management. Early and accurate identification of animal diseases can significantly reduce mortality, economic losses, and prevent the spread of zoonotic infections. Traditional diagnostic methods are time-consuming, require expert veterinarians, and are not always accessible in rural or remote areas.

This project presents a desktop-based application for animal disease prediction using two machine learning techniques — Convolutional Neural Networks (CNN) and Support Vector Machine (SVM). The CNN model is employed for image-based feature extraction and pattern recognition from animal symptoms and affected body regions, while the SVM classifier provides robust classification of disease categories based on extracted features.

The system allows users to upload animal images, which are then preprocessed and passed through the CNN+SVM pipeline to predict the most probable disease condition along with execution time details. Results indicate that the CNN+SVM hybrid approach achieves high prediction accuracy. The paper also discusses future directions including web-based integration, real-time monitoring, and explainable AI for transparent veterinary decision-making.