AI-Driven Dog Care Administration: Integrating Computer Vision and Machine Learning for Intelligent Service Management

Abstract

This article presents a Dog Care Administration System, a web application that integrates pet service management with AI-driven analytics. The system leverages computer vision and machine learning to enhance data integrity and service evaluation. A YOLO-based image verification module authenticates uploaded pet images, reducing erroneous registrations and improving reliability. For customer feedback analysis, machine learning models—including Support Vector Machine, Random Forest, Naïve Bayes, and Neural Networks—automatically classify sentiment to support datadriven service improvement. K-Nearest Neighbors (KNN) is employed as a baseline similarity-based classifier for text tasks, while a refined treestructured approach (KD-Tree/K-Tree) mitigates computational limitations, reducing search complexity and improving scalability for large datasets. Experimental results demonstrate that the enhanced KNN achieves faster inference with competitive accuracy, underscoring the effectiveness of combining computer vision and machine learning for intelligent decisionmaking in modern pet care applications.