基于频繁项集的条件模式挖掘

Mining conditional patterns based on frequent itemsets

在数据库挖掘中,要充分地快速地挖掘出数据库中的任意有趣模式,而现实数据挖掘查询等这种任意合成模式特别复杂,如果只利用现有的基于频繁项集算法直接进行复杂模式挖掘是困难的。为解决该问题,一种基于频繁项集的条件模式挖掘被提出。从条件模式定义,性质,条件模式挖掘算法等方面来阐述解决此类任意条件下模式挖掘的问题。该条件模式的挖掘,使得数据库进行任意模式的新知识新规律发现变得更快捷有效。在现实世界的知识挖掘中,条件模式挖掘更能贴近现实知识的发现。

Mining in the database, it is necessary to quickly and fully mine any interesting patterns in the database, and the reality of data mining, such as for any of this model is particularly complex. If only make use of the existing algorithms based on frequent itemsets, it is difficult to mine directly complex patterns. To solve the problem, mining conditional patterns based on frequent itemsets is presented. From the definition and properties of conditional patterns to conditional pattemsmining algorithm, and other aspects, it is to solve mining issues under such arbitrary conditional patterns. This mode of mining conditions, making arbitrary model of the new knowledge and laws become more efficient and effective. In the real world of knowledge mining, mining conditional patterns more close to the discovery of knowledge in the real world.