增量式知识获取算法综述

Review of incremental learning

粗糙集理论研究中对求解属性的最小约简或较小约简以及求取最简规则集的算法已经进行了一些研究。而数据库是动态的,为了获取最小决策规则集,当增加新数据时,传统的方法通常需要对数据库中所有数据重新计算,效率欠佳,因此对动态数据进行增量式学习是非常必要的。

In recent years, many rough set based algorithms for computing the smallest or smaller reduction of attributes and knowledge acquisition are developed. They are almost based on static data. However, real databases are always dynamic. In order to find a minimal set of rules for a decision table, the classical method cannot effectively deal with new instances added to the universe, because of recalculation for the overall set of instance. Incremental learning of dynamic databases is very necessay.