📄 675.txt
字号:
发信人: easehawking (以鹰行猎), 信区: DataMining
标 题: Re: Decision Tree&Neural Network
发信站: 南京大学小百合站 (Sun Aug 25 01:04:05 2002)
那位帮我下载一下,我在CHINANET访问不了。
easehawking@21cn.com
谢了!!
【 在 helloboy 的大作中提到: 】
: C-Net: A Method for Generating Non-deterministic and Dynamic Multivariate ..
: cision Trees
: Abstract:
: Abstract. Despite the fact that artificial neural networks (ANNs) are univ..
: sal function approximators, their black box nature (that is, their lack of..
: irect interpretability or expressive power) limits their utility. In contr..
: t, univariate decision trees (UDTs) have expressive power, although usuall..
: they are not as accurate as ANNs. We propose an improvement, C-Net, for bo..
: the expressiveness of ANNs and the accuracy of UDTs by consolidating both..
: echnologies for generating multivariate decision trees (MDTs). In addition..
: we introduce a new concept, recurrent decision trees, where C-Net uses rec..
: rent neural networks to generate an MDT with a recurrent feature. That is,..
: memory is associated with each node in the tree with a recursive conditio..
: which replaces the conventional linear one. Furthermore, we show empirical..
: that, in our test cases, our proposed method achieves a balance of compre..
: nsibility and accuracy intermediate between ANNs and UDTs. MDTs are found ..
: be intermediate since they are more expressive than ANNs and more accurat..
: than UDTs. Moreover, in all cases MDTs are more compact (i.e., smaller tre..
: size) than UDTs.
: http://springer.lib.tsinghua.edu.cn/app/home/contribution.asp?wasp=e05d217q
: (以下引言省略...)
--
清除前进路上的一切障碍!
※ 来源:.南京大学小百合站 http://bbs.nju.edu.cn [FROM: 61.140.91.135]
⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -