The previous research on swine disease diagnosis had not applied swine's gender, a significant weight of each
symptom, an occurred symptom certainty factor and major lesion group for the disease diagnosis. Such as expert
system for pre-diagnosis of important swine gastrointestial diseases 2 used symptoms, severe disease level, age span
and feces for diseases diagnosis; the accuracy of diagnosis was 75.4%. A swine's gender, age range, significant
weight of each symptom, an occurred symptom certainty factor and major lesion group were applied in our research
and the result of experiment showed that our system give the disease screening accuracy of 97.50%, the diagnosis by
symptom accuracy of 92.48% and give the diagnosis by lesion accuracy of 95.62% that is higher than the previous
research results. From the result of diagnosis, our system can disease diagnosis approximate to the disease diagnosis
by the veterinarian. And the results of evaluation of satisfaction with Likert-scale by the swine farmers and animal
husbandmen were 4.7 and 4.5 respectively.
Currently, our expert system has been applied in the animal hospital, Faculty of Veterinary Science, Rajamangala
University of Technology Srivijaya and Nakhon Si Thammarat and Songkhla province Livestock Offices. Farmers
and animal husbandman can operate the system through the Internet http://www.swinevet.in.th/swineindex.cfm.