一种改进的知识发现算法及其应用

针对基于留数分析和递归切割的模式发现算法的不足，提出了一种改良的模式发现算法。算法不仅能处理具有特殊分布的样本集，针对不同属性对知识发现的贡献不同而区别对待，同时计及切割产生的子空间数量对模式判别准则的影响。在人造数据集和电力系统安全稳定评估的应用验证了算法的有效性。与原算法相比，知识发现的效率更高、应用范围更广泛。

A Modified Knowledge Discovery Algorithm and Its Application

In view of the limitations of the original pattern discovery algorithm based on residual analysis and recursive partitioning, a modified pattern discovery algorithm is proposed in the paper, which can not only cope with the data set with special distribution effectively but also analyze the different attributes with different treatments during knowledge discovery considering their different contributions, and automatically adjusts the criterion of classifying the subspaces as the patterns according to the variety of the number of subspaces. Test results on synthetic data set and power system stability assessment show its effectiveness. Comparing with the original algorithm, this method is more efficient.