Abstract: Ontology inconsistent problem is one of the key technologies on semantic web.Based on model diagnosis，local diagnosis technology algorithm on E_DAG decribed with classical logic has been widely used in the domain of constructive knowledge description.Innocent diagnosis has improved local diagnosis to obtain more effective result.On the basis of innocent diagnosis，this paper presents an E_DAG representation and its corresponding algorithm，which extends DAG to solve ontology inconsistent problem.An animal ontology example is given to demonstrate the working principle of the E_DAG diagnosis algorithm.

Key words: ontology inconsistent, model diagnosis, local diagnosis, innocent diagnosis, E_DAG

摘要： 本体不一致性问题是语义网络研究中的关键技术之一。基于DAG图的局部诊断方法在传统的模型诊断基础上，采用经典逻辑进行描述，这种方法在构造性知识描述领域中得到大量应用。无辜诊断方法对局部诊断方法进行了改进，可进一步提高局部诊断方法的求解效率。在此基础上，对DAG图表示进行了扩展，提出了一种基于E_DAG的本体不一致性处理的表示方法及其相应实现算法，并通过动物本体实例说明了此方法的工作原理。

关键词: 本体不一致性, 模型诊断, 局部诊断, 无辜诊断, E_DAG