Application And Implement Of Expert System For Ostrich Breeding Information System

Application of MIS(Managing Information System) of large farmingbreeding enterprise is universal and more special in my country because offetching in advanced technology and equipment. But application of MIS ofsmaller farming breeding enterprise on aspect of level of application and scope islate because enterprises develop slowly. Specially professional MIS is infrequent.Professional software on aspect of feedstuff formula and so on is mature and runon PC condition. Ostrich breeding industry is mature although it develops late. Managinginformation take on handwork. Enterprises is blocked because handwork isinefficient. So developing Ostrich MIS is stringent. It is important that ordinary diseases is prevented and cured. Which is shortof software. So software is implemented by ES(Expert System). We adopt UML(Unified Modeling Language)in the course of developingostrich MIS. Which improve rate of developing software by UML. Module of diagnoses on ostrich diseases is based on ES.It includesknowledge base, database, explanation base, knowledge acquisition, interface ofhuman-machine and inference machine. Knowledge base includes the production rule, the reason tree about FebrileDiseases, AND/OR graph in node. Database is composed of descriptions andcases. Explanation base collects the paper copy of the differential diagnosis andreasoning procedure on ostrich diseases. Knowledge acquisition is a learning module. It can modify the productionrule by learning from experts. Interface of human-machine supplies the tools that human communicatewith computer. Inference machine module is the core of system it is a practice programcombined all others module, and includes the production-test-cut procedure ofinference tree about diseases, search strategy of AND/OR graph etc. In order to simplify this problem, we induce and classify various diseases,then repeatedly subdivide to detailed disease. We develop the structure of reasontree about diagnosis of diseases. We make the AND/OR graph in the node of thereason tree. We search the tree through the procedure of production-test-cut inthe search of situational space of artificial intelligence. The production rule 73专家系统在鸵鸟养殖信息系统的应用及实现resulted from the AND/OR graph can be stored to knowledge base.This methodreduces the search cost and is efficient. We also adapt the breadth-first search todetermine evidence of illness.