面向语义检索应用的本体模型结构设计

Novel ontological structure for semantic retrieval

为提高语义检索过程中术语关系自动判别的效果,引入本体理论,对本体的结构设计及推理机理展开研究。提出基于6元组的本体模型结构,采用描述逻辑作为表达工具,实现了对原子术语、复合术语、实例、实例描述、属性分配以及公理的语义描述。依据Gruber设计原则,从明确性和客观性、一致性、最大单向可扩展性、最小编码偏好和最小本体承诺5方面对本体结构进行评价。从语义检索的需求出发,提出面向术语和面向实例的两种推理类型,以及转换机制和判定算法。将语义检索抽象成信息请求与答复过程,提出4类信息搜索策略,并对语义检索系统的实现进行展望。

To enable the automatic identification of the relations between terms in semantic retrieval, ontology structure and inference methods are studied. A six-element based ontological structure is proposed for semantic retrieval. I）eseription logic is adoped as an expression tool to semantically describe the atomic term, complex terms, instances, instances descriptions, attribute assignments and axioms. Then, the new structure is evaluated by the Gruber＇s rules including explicitness, objectivity, consistency, extensibility, minimal encoding bias and minimal ontological commitment. Based on the new structure, this paper proposes two types of reasoning （terms-oriented and instances-oriented） for the application of semantic retrieval, as well as the conversion mechanisms and determining algorithm. Finally, this paper puts forward four kinds of rules for information retrieval, and analyzes the applications of this new structure in semantic retrieval.