Data Classification via Network Method

碩士 === 國立高雄大學 === 統計學研究所博士班 === 104 === The results of classification of traditional data are frequently limited to the choice of the distance. We provide a new concept of converting the traditional data into a network, and then using the network clustering technology to obtain the result of classif...

Full description

Bibliographic Details
Main Authors: Yu-hua Huang, 黃俞華
Other Authors: Kun-Lin Kuo
Format: Others
Language:zh-TW
Published: 2016
Online Access:http://ndltd.ncl.edu.tw/handle/15835881768020796471
id ndltd-TW-104NUK05337002
record_format oai_dc
spelling ndltd-TW-104NUK053370022017-09-17T04:24:05Z http://ndltd.ncl.edu.tw/handle/15835881768020796471 Data Classification via Network Method 透過網絡進行傳統資料分群 Yu-hua Huang 黃俞華 碩士 國立高雄大學 統計學研究所博士班 104 The results of classification of traditional data are frequently limited to the choice of the distance. We provide a new concept of converting the traditional data into a network, and then using the network clustering technology to obtain the result of classification. First, we extend information of each data and treat it as a node of the network. Second, use the multivariate two-sample test method to determine whether two nodes have a link between them. Finally, we cluster this network. The empirical results show that our method has a relatively stable clustering effect. Kun-Lin Kuo 郭錕霖 2016 學位論文 ; thesis 33 zh-TW
collection NDLTD
language zh-TW
format Others
sources NDLTD
description 碩士 === 國立高雄大學 === 統計學研究所博士班 === 104 === The results of classification of traditional data are frequently limited to the choice of the distance. We provide a new concept of converting the traditional data into a network, and then using the network clustering technology to obtain the result of classification. First, we extend information of each data and treat it as a node of the network. Second, use the multivariate two-sample test method to determine whether two nodes have a link between them. Finally, we cluster this network. The empirical results show that our method has a relatively stable clustering effect.
author2 Kun-Lin Kuo
author_facet Kun-Lin Kuo
Yu-hua Huang
黃俞華
author Yu-hua Huang
黃俞華
spellingShingle Yu-hua Huang
黃俞華
Data Classification via Network Method
author_sort Yu-hua Huang
title Data Classification via Network Method
title_short Data Classification via Network Method
title_full Data Classification via Network Method
title_fullStr Data Classification via Network Method
title_full_unstemmed Data Classification via Network Method
title_sort data classification via network method
publishDate 2016
url http://ndltd.ncl.edu.tw/handle/15835881768020796471
work_keys_str_mv AT yuhuahuang dataclassificationvianetworkmethod
AT huángyúhuá dataclassificationvianetworkmethod
AT yuhuahuang tòuguòwǎngluòjìnxíngchuántǒngzīliàofēnqún
AT huángyúhuá tòuguòwǎngluòjìnxíngchuántǒngzīliàofēnqún
_version_ 1718537216004718592