Pet Animal Disease Detection Based on Symptoms

The increasing adoption of pets has created a growing need for intelligent veterinary healthcare systems capable of assisting in early disease diagnosis. Pet animals often suffer from diseases that exhibit overlapping symptoms such as vomiting, coughing, diarrhea, fever, skin infections, and abnormal physiological conditions, making manual diagnosis difficult for non-expert pet owners. This research proposes a machine learning–based Pet Animal Disease Detection System that predicts diseases using symptoms and vital signs. The proposed system utilizes a Random Forest classifier trained on a dataset containing records of multiple animal species including dogs, cats, cows, horses, goats, sheep, pigs, and rabbits. The dataset consists of symptom information, physiological parameters, and disease labels, enabling multi-species disease prediction. Data preprocessing techniques such as cleaning, encoding, and feature alignment were applied before model training. Experimental evaluation demonstrated strong classification performance with 94% accuracy, 93% precision, 92% recall, and 92% F1-score. The system also achieved high reliability in confusion matrix and ROC curve analysis. The proposed approach provides an effective and accessible decision-support tool for early pet disease detection and veterinary healthcare assistance.