Please use this identifier to cite or link to this item:

Title: 以大數據中文詞彙語料庫為基礎的「情緒共現詞彙資料庫」之發展與運用
Authors: 陳彥丞;張雨霖;陳學志;林淑晏;宋曜廷
Chen, Yen-Cheng;Chang, Yu-Lin;Chen, Hsueh-Chih;Lin, Shu-Yen;Sung, Yao-Ting
Contributors: 國立政治大學邁向頂尖大學計畫創新研究團隊
Date: 2016
Issue Date: 2017-06-19 17:31:44 (UTC+8)
Abstract: 情緒與個人的認知、身心健康都有密不可分的關係(陳學志、詹雨臻、馮彥茹,2013),透過情緒文本的分析,有助於了解人類情感狀態及心理健康(Kiefer, Schuch, Schenck, & Fiedler, 2007; St-Hilaire, Cohen, & Docherty, 2008)。世界各國為了研究情緒、分析情緒文本,紛紛建置情緒資料庫,但過去建置情緒資料庫的方式花費許多時間與金錢且詞彙含量少(Jaffe, 2014),因此,本研究擬透過大數據語料庫建置富含大量詞彙的「情緒共現詞彙資料庫」,並依此發展出可用於辨別文本情緒主題的指標。研究一目的在於運用將近43億5千萬個中文詞彙的巨量語料庫,藉由詞彙與情緒種子詞彙集之間的共現性,表徵大量中文詞彙在不同情緒類別上的屬性,並建置「情緒共現詞彙資料庫」。研究二目的為測試研究一的「情緒共現詞彙資料庫」是否能區辨網路文本的情緒類別。The present research aims to build up “the emotional co-lexicon corpus structured by emotional classifications” by using the linguistic big data and develop the Chinese writing indices for identifying the sentimental themes. The purpose of Study 1 is to represent the attributes of a large amount of Chinese lexicons in the different classifications of emotions by using an immense corpus of Chinese lexicons which include about 450 million Chinese words and the lexical collocation. Study 2 aims to test whether we could distinguish the emotional category of article on the Internet by using “the emotional co-lexicon corpus structured by emotional classifications” which is built up in study 1.
Relation: 2016創新研究國際學術研討會: 以人為本的在地創新之跨領域與跨界的對話 2016 International conference on innovation studies- human-centered indigenous innovation: trans-disciplinary dialogue
Data Type: conference
Appears in Collections:[2016創新研究國際學術研討會] 會議論文

Files in This Item:

File Description SizeFormat

All items in 學術集成 are protected by copyright, with all rights reserved.

社群 sharing