Chinese Journal of Computers   Full Text
Title	The Advances in the Covering Based Classification Algorithms
Authors	HE Qing SHI Zhong-Zhi
Address	(Key Laboratory of Intelligent Information Processing,Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100080)
Year	2007
Issue	No.8(1235—1243)
Abstract &   Background	Abstract The understanding of data is highly relevant to how one senses and perceives them. The covering learning algorithms can always simulate human visual cognition to represent the data distribution in the low dimension space. The advances in the area of covering based classification algorithms are summarized. Specially，the Hyper Surface Classification is introduced and analyzed in detail. Moreover, the future research directions are pointed out. keywords covering algorithm; hyper surface classification; minimal consistent subset; machine learning background This paper studies classification problem that belongs to the machine learning category. The advances in the area of classification based on covering algorithm are summarized. Specially, the Hyper Surface Classification is introduced and analyzed in detail. Moreover, the future research directions are pointed out. The understanding of data is highly relevant to how one senses and perceives them. The covering learning algorithms can always simulate human visual cognition to represent the data distribution. Some covering learning algorithms were proposed in the decade. Zhang Ling and Zhang Bo proposed a geometry classification method, where the original input space is transferred into a quadratic space by the use of a global project function. Then, the well-known point set covering method was used to perform the partition of the data in the transformed space. Biomimetic Pattern Recognition(BPR) theory is firstly proposed by Wang Shou-Jue as a new model for pattern recognition. Xu Zong-Ben proposed an approach called Visual Classification Algorithm(VCA) for classification, with an expectation of resolving some of the problems mentioned above. Lee Daewon and Lee Jaewook proposed a learning algorithm for semi-supervised classification. For Hyper Surface Classification(HSC), Hyper Surface Classification(HSC) is a novel classification method based on hyper surface is put forward by He & Shi & Ren(2002). However, what we really need is an algorithm that can deal with data not only of massive size but also of high dimensionality. Thus He, Zhao & Shi proposed a simple and effective kind of dimension reduction method without losing any essential information in 2006. Another solution to the problem of HSC on high dimensional data sets is proposed. A judgment sampling method based on Minimal Consistent Subset(MCS) is proposed to select of a representative subset of the original training data.