An Application of Multi-Class Online Transfer Learning on 4G/LTE Network Traffic Data Analysis

碩士 === 國立交通大學 === 統計學研究所 === 106 === Propose another framework of Online Transfer Learning that can solve the multi-class classification task. To transfer the knowledge of a source domain to a target domain, we combine the source classifier and the online target classifier by allocating different we...

Full description

Bibliographic Details
Main Authors:	Hung, Hsiu-Chun, 洪修淳
Other Authors:	Lu, Horng-Shing
Format:	Others
Language:	en_US
Published:	2018
Online Access:	http://ndltd.ncl.edu.tw/handle/aq8xbs

id	ndltd-TW-106NCTU5337027
record_format	oai_dc
spelling	ndltd-TW-106NCTU53370272019-05-16T01:24:32Z http://ndltd.ncl.edu.tw/handle/aq8xbs An Application of Multi-Class Online Transfer Learning on 4G/LTE Network Traffic Data Analysis 多類別線上轉移學習在4G/LTE網路流量資料分析應用 Hung, Hsiu-Chun 洪修淳碩士國立交通大學統計學研究所 106 Propose another framework of Online Transfer Learning that can solve the multi-class classification task. To transfer the knowledge of a source domain to a target domain, we combine the source classifier and the online target classifier by allocating different weights. We introduce a concept of possibility vector and combine the possibility vectors of two classifiers to make the prediction. Then we develop a new mechanism for updating these allocation weights. We also provide theoretical analysis to guarantee the performance of the framework will not be too bad. At last we apply the framework to analyze 4G/LTE network traffic data. Lu, Horng-Shing 盧鴻興 2018 學位論文 ; thesis 28 en_US
collection	NDLTD
language	en_US
format	Others
sources	NDLTD
description	碩士 === 國立交通大學 === 統計學研究所 === 106 === Propose another framework of Online Transfer Learning that can solve the multi-class classification task. To transfer the knowledge of a source domain to a target domain, we combine the source classifier and the online target classifier by allocating different weights. We introduce a concept of possibility vector and combine the possibility vectors of two classifiers to make the prediction. Then we develop a new mechanism for updating these allocation weights. We also provide theoretical analysis to guarantee the performance of the framework will not be too bad. At last we apply the framework to analyze 4G/LTE network traffic data.
author2	Lu, Horng-Shing
author_facet	Lu, Horng-Shing Hung, Hsiu-Chun 洪修淳
author	Hung, Hsiu-Chun 洪修淳
spellingShingle	Hung, Hsiu-Chun 洪修淳 An Application of Multi-Class Online Transfer Learning on 4G/LTE Network Traffic Data Analysis
author_sort	Hung, Hsiu-Chun
title	An Application of Multi-Class Online Transfer Learning on 4G/LTE Network Traffic Data Analysis
title_short	An Application of Multi-Class Online Transfer Learning on 4G/LTE Network Traffic Data Analysis
title_full	An Application of Multi-Class Online Transfer Learning on 4G/LTE Network Traffic Data Analysis
title_fullStr	An Application of Multi-Class Online Transfer Learning on 4G/LTE Network Traffic Data Analysis
title_full_unstemmed	An Application of Multi-Class Online Transfer Learning on 4G/LTE Network Traffic Data Analysis
title_sort	application of multi-class online transfer learning on 4g/lte network traffic data analysis
publishDate	2018
url	http://ndltd.ncl.edu.tw/handle/aq8xbs
work_keys_str_mv	AT hunghsiuchun anapplicationofmulticlassonlinetransferlearningon4gltenetworktrafficdataanalysis AT hóngxiūchún anapplicationofmulticlassonlinetransferlearningon4gltenetworktrafficdataanalysis AT hunghsiuchun duōlèibiéxiànshàngzhuǎnyíxuéxízài4gltewǎnglùliúliàngzīliàofēnxīyīngyòng AT hóngxiūchún duōlèibiéxiànshàngzhuǎnyíxuéxízài4gltewǎnglùliúliàngzīliàofēnxīyīngyòng AT hunghsiuchun applicationofmulticlassonlinetransferlearningon4gltenetworktrafficdataanalysis AT hóngxiūchún applicationofmulticlassonlinetransferlearningon4gltenetworktrafficdataanalysis
_version_	1719175832048500736

An Application of Multi-Class Online Transfer Learning on 4G/LTE Network Traffic Data Analysis

Similar Items