國立高雄大學圖資館 |

Language: English

Back

應用自我組織圖於社會網路文字訊息之情感分析 = Sentiment An...

吳俊諺

應用自我組織圖於社會網路文字訊息之情感分析 = Sentiment Analysis of Text Messages in Social Networks Based on Self-Organizing Maps

Record Type:	Language materials, printed : monographic
Paralel Title:	Sentiment Analysis of Text Messages in Social Networks Based on Self-Organizing Maps
Author:	吳俊諺,
Secondary Intellectual Responsibility:	國立高雄大學
Place of Publication:	[高雄市]
Published:	撰者;
Year of Publication:	2013[民102]
Description:	56面圖，表格 : 30公分;
Subject:	情感分析
Subject:	Text Mining
Online resource:	http://handle.ncl.edu.tw/11296/ndltd/40097979979130181291
Notes:	參考書目：面45-49
Notes:	102年10月31日公開
Summary:	隨著網際網路的發展和普及，以文本形式出現的訊息量也急遽地增長，對於其上資料分析的需求量也隨之增加。透過情感分析可以從大量的文字內容中發掘出具價值的知識，可得到可觀的商業、政治、經濟、情報、國安等利益，為企業或個人提供意見或喜好資訊，以支援決策者進行最佳化決策。然而社會網路的文字訊息不同於一般文字文件之特性，在進行文本探勘時便具有其差異性與困難度，故如何在社會網路的環境下，發展一套有效能的概念探勘之方法是不可或缺的。本研究將採用類神經網路中文件分群之方法，即自我組織圖，將不同的訊息與情感概念加以分群，最後利用相似度分析來協助我們找尋訊息與概念間之關聯。針對所發掘之關聯結果，本研究開發適合社會網路文字訊息之探勘技術，以偵測出訊息之情感概念。 With explosive growth of the Internet, the amount of information in text form is growing rapidly and the demand for data analysis is also increasing. We can perform sentiment analysis on a large set of text messages to discover valuable knowledge and obtain enormous benefits in national security, business, politics, economics, , etc, However, text messages from the social networks are rather different from those of traditional text documents. Therefore, it is difficult but essential to develop an effective method of sentiment exploration in social networks. In this study we use a neural network method for document clustering, namely the self-organizing maps. We first applied self-organizing maps to cluster similar messages and sentiment keywords. We then developed an association discovery process to find the associations between the messages and sentiment keywords. The sentiment of a message is then determined according to such associations. We performed experiments on Twitter messages and obtained promising results.

應用自我組織圖於社會網路文字訊息之情感分析 = Sentiment Analysis of Text Messages in Social Networks Based on Self-Organizing Maps
吳, 俊諺

應用自我組織圖於社會網路文字訊息之情感分析 = Sentiment Analysis of Text Messages in Social Networks Based on Self-Organizing Maps / 吳俊諺撰 - [高雄市] : 撰者, 2013[民102]. - 56面 ; 圖，表格 ; 30公分.
參考書目：面45-49102年10月31日公開.
情感分析Text Mining

應用自我組織圖於社會網路文字訊息之情感分析 = Sentiment Analysis of Text Messages in Social Networks Based on Self-Organizing Maps
LDR:03165nam0a2200289 450 001 389703
005 20170214100947.0
009 389703
010 0 $b 精裝
010 0 $b 平裝
100 $a 20170214d2013 k y0chiy05 e
101 1 $a chi $d chi $d eng
102 $a tw
105 $a ak am 000yy
200 1 $a 應用自我組織圖於社會網路文字訊息之情感分析 $d Sentiment Analysis of Text Messages in Social Networks Based on Self-Organizing Maps $z eng $f 吳俊諺撰
210 $a [高雄市] $c 撰者 $d 2013[民102]
215 0 $a 56面 $c 圖，表格 $d 30公分
300 $a 參考書目：面45-49
300 $a 102年10月31日公開
314 $a 指導教授：楊新章博士
328 $a 碩士論文--國立高雄大學資訊管理學系碩士班
330 $a 隨著網際網路的發展和普及，以文本形式出現的訊息量也急遽地增長，對於其上資料分析的需求量也隨之增加。透過情感分析可以從大量的文字內容中發掘出具價值的知識，可得到可觀的商業、政治、經濟、情報、國安等利益，為企業或個人提供意見或喜好資訊，以支援決策者進行最佳化決策。然而社會網路的文字訊息不同於一般文字文件之特性，在進行文本探勘時便具有其差異性與困難度，故如何在社會網路的環境下，發展一套有效能的概念探勘之方法是不可或缺的。本研究將採用類神經網路中文件分群之方法，即自我組織圖，將不同的訊息與情感概念加以分群，最後利用相似度分析來協助我們找尋訊息與概念間之關聯。針對所發掘之關聯結果，本研究開發適合社會網路文字訊息之探勘技術，以偵測出訊息之情感概念。 With explosive growth of the Internet, the amount of information in text form is growing rapidly and the demand for data analysis is also increasing. We can perform sentiment analysis on a large set of text messages to discover valuable knowledge and obtain enormous benefits in national security, business, politics, economics, , etc, However, text messages from the social networks are rather different from those of traditional text documents. Therefore, it is difficult but essential to develop an effective method of sentiment exploration in social networks. In this study we use a neural network method for document clustering, namely the self-organizing maps. We first applied self-organizing maps to cluster similar messages and sentiment keywords. We then developed an association discovery process to find the associations between the messages and sentiment keywords. The sentiment of a message is then determined according to such associations. We performed experiments on Twitter messages and obtained promising results.
510 1 $a Sentiment Analysis of Text Messages in Social Networks Based on Self-Organizing Maps $z eng
610 0 $a 情感分析 $a 文本探勘 $a 自我組織圖
610 1 $a Text Mining $a Sentiment Analysis $a Self-Organizing Map
681 $a 008M/0019 $b 464105 2620 $v 2007年版
700 1 $a 吳 $b 俊諺 $4 撰 $3 614563
712 0 2 $a 國立高雄大學 $b 資訊管理學系碩士班 $3 353936
801 0 $a tw $b NUK $c 20131004 $g CCR
856 7 $z 電子資源 $2 http $u http://handle.ncl.edu.tw/11296/ndltd/40097979979130181291