论文部分内容阅读
定量化的土壤系统分类体系的建立推动了土壤类型自动检索的研究。但以往研究并未关注土壤系统分类体系仍需不断修订的现状,所开发系统的推理代码与知识规则通常互相绑定,当知识规则改变时,推理代码需要重新开发,导致已有的系统被弃用,这阻碍了土壤类型自动检索系统的推广和应用。针对上述问题,以《中国土壤系统分类检索》(第三版)为基础,建立了基于RuleML语言的、可随时修改和扩展的规则库。然后利用规则引擎工具NxBRE,使土壤类型的检索规则与推理过程相分离,知识规则的改变不需要重新编写推理过程代码,并最终实现了土壤类型的自动检索。
The establishment of a quantitative soil system classification system has promoted the research of soil type automatic retrieval. However, previous researches did not pay attention to the status quo that the soil system classification system needs to be revised continuously. The reasoning codes and knowledge rules of the developed system are usually bound to each other. When the knowledge rules change, the inference code needs to be re-developed, resulting in the abandonment of the existing system This hindered the popularization and application of soil type automatic retrieval system. In response to the above problems, based on “China soil system classification search” (third edition), based on the RuleML language, the rule base can be modified and extended at any time. Then using the rule engine tool NxBRE, the soil type retrieval rules are separated from the reasoning process. The knowledge rules change does not need to rewrite the inference process code, and finally the soil type is automatically retrieved.