Design of fertilization recommendation knowledge base and appllication

Crop fertilization recommendation system involves using models to calculate the needed amount of variety of nutrients during the crop growth, choosing suitable fertilizers, and arranging fertilization time. Whether it can be used widely or not, the key point is that the models or parameters in system can be customized easily according with local agricultural production practices. To help address these issues, an infrastructure of knowledge base and its application is proposed. This paper firstly focuses on decomposition of the model by method of object-oriented in order to comply with the requirements of C++ programming. It is divided into three categories of entity, parameter, and operator for converting the entity objects in fertilization to the software system objects. And then the required knowledge to run model are classified to four types by their action, and expressed as a variety of rules form stored in relation database. In the end, a reasonable decision inference engine designed for applying them. It is actually a specific computer program to control local entities and the rules form introduced to system under a certain strategy and produce appliable recommendation.