Abstract: The authors define an event and an event class, and illustrate the relations between event classes. Then event-oriented ontology model is constructed, which will contribute new methods and technologies to semantic-based knowledge processing. Furthermore, taking event class ranking as an example based on event ontology, the method of HARank-based event class ranking is introduced for event ontology, and corresponding experimental and evaluating results are presented. In comparison with conventional methods, event ontology represents knowledge with a higher granularity, and it will be more suitable for the representation of the knowledge about the movement and state changes in the real world.

Key words: event, event class, event class relation, event ontology, event class ranking

摘要： 给出了事件、事件类的定义, 分析了事件类之间的关系, 在此基础上形成了面向事件的本体模型, 为基于语义的知识处理提供了新的方法与技术。在事件本体模型的基础上以事件类排序为实例, 研究了基于HARank (Hubs-Authorities Rank)算法的事件类排序方法, 给出了实验结果, 并对其进行了评价。与传统本体相比, 事件本体以更高粒度的事件作为知识表示单元, 更适合表示运动世界的知识。

关键词: 事件类, 事件类关系, 事件本体, 事件类排序, 事件