%0 Journal Article
%T COCA - a new way to auto-detect association based on entropy correlated coefficients<br>基于熵相关系数的关联性自动判别方法——COCA
%A WANG Shan
%A CAO Wei
%A QIN Xiong-pai
%A <br>王珊
%A 曹巍
%A 覃雄派
%J 计算机应用
%D 2006
%I 
%X Self-managing and self-optimizing is currently a hot research field in database.To guarantee the accuracy of the estimates made by optimizer,this paper proposed a new method named COCA(entropy-COrrelated-Coefficient-based Auto-detection of association).In comparison with CORDS,COCA has the following features:(1) Fewer limitations.It overcomes the limitation that Chi-square test needs at least 80% of the cells in the contingency table have frequencies greater than 5.(2) More results.CORDS can tell the correlation between columns,while COCA can further discern the specific association degree for both directions.Experiments show that COCA is more robust and produces more statistical information,which is supportive to the creation of more effective and efficient histograms.
%K query optimization
%K statistical information
%K correlation
%K entropy correlated coefficient<br>查询优化
%K 统计信息
%K 关联性
%K 熵相关系数
%U http://www.alljournals.cn/get_abstract_url.aspx?pcid=5B3AB970F71A803DEACDC0559115BFCF0A068CD97DD29835&cid=8240383F08CE46C8B05036380D75B607&jid=831E194C147C78FAAFCC50BC7ADD1732&aid=B9E61A5388FE0A86&yid=37904DC365DD7266&vid=96C778EE049EE47D&iid=9CF7A0430CBB2DFD&sid=2DD7160C83D0ACED&eid=67289AFF6305E306&journal_id=1001-9081&journal_name=计算机应用&referenced_num=1&reference_num=4