摘要
数据挖掘主要用于从原始数据资料中挖掘有用的信息,而这些数据资料的维数已经对目前大多数数据挖掘算法的效率造成了严重的阻碍,这种阻碍被称之为"维数灾难"。数据降维技术可以有效地解决这一问题。文章以数据降维方法为主线,对数据降维问题的分类进行了描述,对数据降维方法的研究现状及主要算法进行了详细的阐述,对数据降维算法最新研究进展进行了简要介绍,并指出其优缺点,最后提出了数据降维技术今后的研究方向。
Data mining is mainly used for the mining of useful information from raw data, however the dimensions of the raw data have become a serious obstacle to the efficiency of the most data mining algorithms. The obstacle is called as the ＂dimensional disaster＂ . The data dimension reduction technology can be used to effectively solve this problem. Taking the data dimension reduc- tion method as the main clue, this article describes the classification of data dimension reduction, expatiates on the research status and main algorithms of the data dimension reduction method, gives a brief description of the latest research progress on data dimen- sion reduction algorithms, and points out the advantages and disadvantages. Finally, the article presents the future research direc- tion of data dimension reduction technology.
出处
《情报理论与实践》
CSSCI
北大核心
2013年第2期125-128,共4页
Information Studies:Theory & Application
基金
吉林大学"985工程"项目资助的研究成果
关键词
数据挖掘
高维数据
降维
研究现状
研究进展
data mining
high dimension data
dimension reduction
research status
research progress
