淡江大學機構典藏:Item 987654321/87732
English  |  正體中文  |  简体中文  |  Items with full text/Total items : 62822/95882 (66%)
Visitors : 4018731      Online Users : 1067
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library & TKU 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
    Home ‧ Login ‧ Upload ‧ Help ‧ About ‧ Administer Goto mobile version


    Please use this identifier to cite or link to this item: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/87732


    Title: 部落格文章情感分析之研究
    Other Titles: A study of sentimental analysis in blog articles
    Authors: 簡之文;Chien, Chih-Wen
    Contributors: 淡江大學資訊管理學系碩士班
    蕭瑞祥
    Keywords: 情感分析;支援向量機;知網;主觀情緒;sentimental analysis;SVM;Hownet;subjective emotion
    Date: 2012
    Issue Date: 2013-04-13 11:36:49 (UTC+8)
    Abstract: 如何在龐大的網路社群文章中，有效且快速的擷取所要的情感評論，是情感分析(Sentiment Analysis)重要的基礎工作，本研究選擇在語句層級探討，分析文章中的主觀情緒評論語句。嘗試找出主觀情緒語句與非主觀情緒語句的判斷模式。本研究採用系統發展研究法，使用SVM(Support Vector Machine)工具將主觀情緒語句進行訓練與測試，實驗中再將SVM產出的分類與實際分類做比對，計算其準確率做為系統驗證的依據。
    根據本研究的實驗發現，依據SVM工具產出的分類，若將屬性詞與意見片語之距離容許誤差值設為3，可大幅提升距離分類準確率，且可減少距離比對的運算成本，而屬性詞類的分類特徵較能區分主觀情緒語句與意見片語距離的關係，與相關研究中的演算法比較後，可將過濾非主觀情緒階段的準確率提升13%，達到66%。期望本研究能為部落格，甚至其他社群網路情感分析之研究與實務有所貢獻。
    How to capture the emotional opinion efficiently and quickly in a lot of network community articles is an important basic work for Sentiment Analysis. This study is discussed in the sentence-level analysis of the subjective emotional comment sentences in the article, and tries to find out a judgment model of subjective emotional sentences and non-subjective emotional sentences. This study adopts Systems Development Methodology, and it uses SVM tool to train and test the subjective emotional sentences. In this experiment, it compares the classification of SVM with the actual classification and it calculates the accuracy as the basis for system verification.
    On the basis of the classification of SVM, this study found if the tolerance value of the distance between attribute word and opinion phrase is set to 3, it will significantly improve the accuracy of distance classification and reduce the computing cost of distance comparison, and then the classification feature of the attribute words is better than negative words and opinion words in recognizing the distance relationship between attribute word and opinion phrase.
    Appears in Collections:[Graduate Institute & Department of Information Management] Thesis

    Files in This Item:

    File SizeFormat
    index.html0KbHTML269View/Open

    All items in 機構典藏 are protected by copyright, with all rights reserved.

    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library & TKU Library IR teams. Copyright ©   - Feedback