一种基于系统不确定性的主动式知识学习算法

An initiative knowledge learning algorithm based on system uncertainty

主动式知识学习方法不依赖于先验领域知识,其学习过程受信息系统的内在属性控制,所产生的知识系统能够更客观地表达信息系统的潜在特征和规律。系统不确定性能够有效地控制主动式知识学习过程。基于粗糙集理论,提出一种系统不确定性度量方式,结合Skowron算法,设计出一种基于系统不确定性的主动式决策规则知识学习算法。仿真实验结果表明该算法能够更好地适应系统的不确定性,其综合性能明显优于其他同类算法。

The initiative knowledge learning method is independent of prior domain knowledge. Its learning processes can be controlled by some inherent features of information systems. Thus, its generated knowledge systems can more objectively express the potential characteristics and patterns of information systems. Initiative learning method can be effectively implemented based on system uncertainty. Herein, a measurement of system uncertainty is developed based on rough set theory; then a new initiative knowledge learning algorithm is proposed on the basis of the new system uncertainty measurement and the Skowron＇s algorithm for mining propositional default decision rules. As is illustrated by simulation experiments, the new algorithm can be more adaptable to system uncertainty; its comprehensive performances are observably better than those of similar algorithms.