文档介绍:Dynamic Partial Coverage Based Feature Selection Method Yu Huang 1,2, Gongde Guo 1,2 , Tianqiang Huang 1,2 and Hong Chen 1,2 1 School of Mathematics puter Science, Fujian Normal University 2 Key Laboratory work Security and Cryptography ,Fujian Normal University Fuzhou, Fujian, China, 350007 yellowfish2001@, ******@fjnu. Abstract In this paper, we propose a novel feature selection method based on spatial coverage relations of features in multidimensional data space. As a filter solution, the algorithm can evaluate the weight of each feature by calculating the spatial coverage relations of features of instances with the same and different class labels in multidimensional data space. And the approach is simple to implement. The experimental results evaluated on some public data set downloaded from the UCI machine learning repository show that the proposed pares well with some classical feature selection methods such as Relief and SVMAttributeEval which are implemented in Weka. 1. Introduction In real-world applications, . in text mining or in predictive toxicology, the number of features in a dataset maybe very large as it could include many irrelevant or redundant features. Most of learning algorithms could not work well with the high dimensional data. This is so-called “curse of dimensionality”[1]. It is necessary to reduce the number of features in or