P-信息隐藏挖掘的数量特征及应用

经典的集合具有静态特性。在科技迅猛发展的今天,海量数据变化更新之迅速,用经典集合来研究数据挖掘问题受到了限制,需要把动态特性引入到普通集合中,用“动态性”代替普通集合的“静态性”,而P-集合具有上述特征。利用P-集合与它的属性迁移,提出内、外k阶P-信息的粒度、挖掘度,讨论了属性迁移与信息挖掘的数量关系;给出迁移信息链式定理及最小、最大挖掘度定理。P-信息是动态信息系统知识挖掘的一个新的理论与方法,最后给出P-信息在数据挖掘中的应用。

Quantitative characteristics and applications of P-information hidden mining

The classical set has static characteristics. With the rapid development of science and technology, massive data updates very quickly. Therefore using the classical set to study data mining problems has been limited. It is necessary to involve dynamic characteristics into ordinary set and replace "static" by "dynamic". Thus P-set has the above characteristics. By using P-set and its attribute migration, the granularity and mining degree of internal and external k-order P-information are proposed; also the quantitative relationship between attribute migration and information mining is discussed. The migration information chain theorem and the minimum and maximum mining degree theorem are given. P-information is a new theory and method for dynamic information system knowledge mining. Finally, the application of P-information in data mining is given.