Comparison of Accuracy Level of Certainty Factor Method and Bayes Theorem on Cattle Disease

This study aims to address the challenges of livestock disease diagnosis in Okaba district, Meraoke, Papua. A total of 2 paramedics or veterinarians and 1 assistant is not sufficient because of the long distances that the medics have to travel, traveling from all areas of Okaba District to its interior. Keepers can only utilize their basic skills for temporary care. The researcher's process included interviews with experts covering the disease, its symptoms and prevention, then analyzed with the provision of utilizing certainty factors and Bayes' theorem to increase the accuracy and veracity of the findings. In this scenario, the data is used as a reference point for analysis in the web-based expert system. The results obtained when processing the problem estimation are disease information, symptom information, and treatment. The reference in the application and analysis shows that the Certainty Factor method is superior in providing consistent accuracy, with a percentage reaching 98.79% in the case of worms, while the Bayes Theorem method shows lower accuracy, around 73%. The comparison indicates that Certainty Factor is more suitable in high uncertainty environments, while Bayes' Theorem is more effective when sufficient probabilistic data is available. Future suggestions can expand the scope by testing other methods such as Machine Learning or Artificial Neural Networks to increase the accuracy of the diagnosis percentage. In addition, more extensive trials on different types of livestock and different environmental conditions will help in developing a more flexible and robust system.