English  |  正體中文  |  简体中文  |  Items with full text/Total items : 87250/116256 (75%)
Visitors : 23281766      Online Users : 272
RC Version 6.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
Scope Tips:
  • please add "double quotation mark" for query phrases to get precise results
  • please goto advance search for comprehansive author search
  • Adv. Search
    HomeLoginUploadHelpAboutAdminister Goto mobile version
    政大機構典藏 > 商學院 > 統計學系 > 學位論文 >  Item 140.119/122378
    Please use this identifier to cite or link to this item: http://nccur.lib.nccu.edu.tw/handle/140.119/122378


    Title: 基於Word2Vec台灣媒體政治傾向探討
    A Study of word2vec: Application of Media Bias Investigation
    Authors: 黃中廷
    Huang, Chung-Ting
    Contributors: 薛慧敏
    黃中廷
    Huang, Chung-Ting
    Keywords: 媒體偏差
    政治新聞
    Word2Vec
    相關性
    Media bias
    Political news
    Word2Vec
    Correlation
    Date: 2019
    Issue Date: 2019-03-04 19:12:14 (UTC+8)
    Abstract:   新聞媒體在資訊傳遞與監督政府上扮演重要的角色,但媒體偏差的問題也伴隨著大量的報導一同產生,尤其是政治新聞。近期Word2vec方法被用來將類別型態的字詞映射至實數向量空間上,經過量化後,字詞間的相關性得以被測量。本論文將Word2Vec應用於台灣電子媒體之新聞資料,透過提取關鍵字的方式加以分析,以探討媒體偏差之存在性。我們並探究Word2Vec中的模型與傳統統計觀點異同之處。本研究使用2014年台灣地區的政治類新聞,比較兩家電子媒體對台北市長候選人相關新聞,另外也檢測word2vec方法中的窗格大小對分析結果的敏感度。我們發現不同的媒體在用字與寫作風格上有明顯差異,另外也存在著政黨偏好的可能性。
      News media plays an important role in information transmission and supervising the government, but the problem of media bias is accompanied with massive numbers of news especially in political news. Word2Vec is used to map categorical data into real number space. The correlation between words can be measured after quantifying. In this paper, we apply Word2Vec on the news data of Taiwan electronic media, capturing keywords and analyzing them to find out the existence of media bias. We also explore the differences of views between the model in Word2Vec and original statistics concepts. In this research, we use the political news in Taiwan in 2014, comparing news associated with candidates of the mayor in Taipei. On the other hand, we do some detection on the sensitivity of the window size to Word2Vec. Finally we discover that besides different media having different writing habit themselves, they also have the possibility of party preferences.
    Reference: [1] 台北市媒體服務代理商協會,(2017).2017年台灣媒體白皮書.
    https://maataipei.org/download/2017媒體白皮書/
    [2] D'Alessio, D, Allen, M,(2000).Media Bias in Presidential
    Elections: A Meta-Analysis, Journal of Communication,50,133-156.
    [3] 媒體改造學社、台灣媒體觀察教育基金會、優質新聞發展協會,(2018).
    【聯合聲明】選舉新聞嚴重失衡,媒體自律形同具文,媒體觀察組織發表
    嚴厲譴責暨申訴行動聲明.http://www.mediawatch.org.tw/news/9787。
    [4] Eberl J.-M., Boomgaarden H.G., Wagner M.,(2017).One Bias Fits
    All? Three Types of Media Bias and Their Effects on Party
    Preferences, Communication Research,44,1125-1148.
    [5] Sun J.Y., 结巴中文分词.https://github.com/fxsjy/jieba
    [6] Cavnar W.B., Trenkle J.M.,(1994). N-Gram-Based Text
    Categorization, Proceedings of SDAIR-94, 3rd Annual Symposium on
    Document Analysis and Information Retrieval,161-175.
    [7] Pennington J., Socher R., Manning C.D.,(2014). GloVe: Global
    Vectors for Word Representation, EMNLP 2014,1532-1543.
    [8] Selivanov D.,(2015). GloVe vs word2vec revisited,
    http://dsnotes.com/post/glove-enwiki/
    [9] McCormick C.,(2016). Word2Vec Tutorial-The Skip-Gram Model.
    http://mccormickml.com/2016/04/19/word2vec-tutorial-the-skip-
    gram-model/
    [10] Mikolov T., Chen K., corrado G.S., Dean J.,(2013). Efficient
    Estimation of Word Representations in Vector Space,
    arXiv:1301.3781v3.
    [11] Mikolov T., Sutskever I., Chen K., Corrado G.S., Dean J.,(2013).
    Distributed Representations of Words and Phrases and their
    Compositionality, NIPS 2013,3111-3119.
    Description: 碩士
    國立政治大學
    統計學系
    105354013
    Source URI: http://thesis.lib.nccu.edu.tw/record/#G0105354013
    Data Type: thesis
    DOI: 10.6814/THE.NCCU.STAT.003.2019.B03
    Appears in Collections:[統計學系] 學位論文

    Files in This Item:

    File SizeFormat
    401301.pdf1416KbAdobe PDF0View/Open


    All items in 政大典藏 are protected by copyright, with all rights reserved.


    社群 sharing

    著作權政策宣告
    1.本網站之數位內容為國立政治大學所收錄之機構典藏,無償提供學術研究與公眾教育等公益性使用,惟仍請適度,合理使用本網站之內容,以尊重著作權人之權益。商業上之利用,則請先取得著作權人之授權。
    2.本網站之製作,已盡力防止侵害著作權人之權益,如仍發現本網站之數位內容有侵害著作權人權益情事者,請權利人通知本網站維護人員(nccur@nccu.edu.tw),維護人員將立即採取移除該數位著作等補救措施。
    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - Feedback