171.txt

来自「This complete matlab for neural network」· 文本 代码 · 共 26 行

发信人: fervvac (高远), 信区: DataMining
标  题: Re: 几个简单的问题！
发信站: 南京大学小百合站 (Sat Jun 29 00:49:28 2002), 站内信件

You seem to already have that paper ya, just read it carefully, it is at least
an VLDB paper, :)

The definition of metric space and similarity search should already been 
defined in the paper. Basically, you can imagine a metric space being a
n-D space, where there is no (total) order for any dimension. For example,
color. But there is a mechanism to define similarity measure.

【 在 mycat (cat) 的大作中提到: 】
: 有几个单词,句子不知如何翻译：
: metric space;
: multi-dimensional access methods ,such as R-tree,have been considered to ind
: ex such data.
: access method;
: indexing a metric space means to provide an efficient support for answering 
: similarity queries
: 还有，有谁知道m-tree吗，能简单介绍一下它的机制嘛？它和R-tree类似！


--
※ 来源:．南京大学小百合站 bbs.nju.edu.cn．[FROM: 饮水思源BBS]