This study developed a web-based expert system for the early diagnosis of canine skin diseases, utilizing the Dempster-Shafer method. The close bond between humans and dogs, the high rate of dog ownership, and the prevalence of skin diseases among dogs in Bali Province highlight the need for an effective solution, particularly given dog owners' limited knowledge of diagnosis and treatment options. The research involved interviews with veterinary experts and an extensive literature review, resulting in the creation of a knowledge base comprising 12 common canine skin diseases and 23 associated symptoms. The Dempster-Shafer method was employed to address uncertainty in decision-making by calculating belief and plausibility values, ensuring accurate diagnostic results. The system's database design enables efficient management of disease, symptom, patient, and diagnosis information. Testing demonstrated an accuracy rate of 86.36% and a user satisfaction score of 89.09%, indicating that the system is both reliable and user-friendly. This expert system provides a practical tool for dog owners in Bali, supporting early diagnosis and appropriate management of canine skin diseases.