摘要
粗糙集理论为研究不精确数据的分析、推理,挖掘数据间的关系、发现潜在的知识提供了有效的工具。在数据挖掘技术中KNN算法是一个实现简单和分类准确性较高的方法,但是,当用于样本容量较大以及特征属性较多的类似医疗图像挖掘这样的领域时,其效率受到了很大的影响,找到一个删除最大冗余属性的方法成了解决这个问题的关键。将粗糙集理论与KNN算法结合起来,用粗糙集方法进行属性约简,有效地解决了KNN算法分类的这个缺点。
Rough Set Theory is an effective tool used to analyze imprecise data and discover the latent rules in these data. While KNN algorithm is a classical algorithm in classification areas. It has been widely used in many areas due to its simplicity and its classification accuracy, but when there is larger capacity and more attributes of the samples, the classification efficiency of the algorithm is affected badly, such as in Medical images mining area . So the key to solve the problem is to find an approach which can be used to delete the redundancy attributes of the samples. In order to improve this defect efficiently, a rough set theory based on KNN algorithm is proposed.
出处
《广西师范大学学报（自然科学版）》
CAS
北大核心
2007年第4期75-78,共4页
Journal of Guangxi Normal University:Natural Science Edition
基金
贵州省自然科学基金资助项目(黔科合J字[2005]2106号)
关键词
KNN分类
粗糙集
属性约简
KNN classification rough set attributes reduction
