Abstract The literature on inconsistency measures has ignored a distinction, that is, differentiating absolute measures and relative measures. An absolute measure gives the total amount of inconsistency in the knowledge base but a relative measure computes, by some criteria, the proportion of the base that is inconsistent. To compare the inconsistency measures, researchers have proposed postulates for such measures. We split these postulates into three groups: ones (including two new postulates) that relative measures should satisfy, ones inappropriate for relative measures, and ones that relative measures may satisfy. We obtain some new results upon the relationships between these groups of postulates. On these grounds, we introduce a formal definition for relative inconsistency measures. We consider some relative measures previously proposed and define several new ones that serve as examples. We show that all of these measures satisfy the new formal definition.

中文翻译：

---

相对不一致性度量

摘要 关于不一致测度的文献忽略了一个区别，即区分绝对测度和相对测度。绝对度量给出了知识库中不一致性的总量，而相对度量根据某些标准计算了不一致的基数比例。为了比较不一致的措施，研究人员提出了这些措施的假设。我们将这些假设分为三组：相对测度应该满足的假设（包括两个新假设），不适合相对测度的假设，以及相对测度可能满足的假设。我们在这些假设组之间的关系上获得了一些新的结果。基于这些理由，我们引入了相对不一致度量的正式定义。我们考虑了之前提出的一些相关措施，并定义了几个作为示例的新措施。我们表明所有这些措施都满足新的正式定义。