簡易檢索 / 詳目顯示

回結果列表
研究生： 賴建江
Lai, Jian-Jiang
論文名稱： 基於人際關係知識圖之新聞名人聊天機器人
A News Celebrity Chatbot Based on Interpersonal Relationship Knowledge Graph
指導教授： 盧文祥
Lu, Wen-Hsiang,
學位類別： 碩士
Master
系所名稱： 電機資訊學院 - 資訊工程學系
Department of Computer Science and Information Engineering
論文出版年： 2020
畢業學年度： 108
語文別： 英文
論文頁數： 33
中文關鍵詞： 人際關係抽取 、知識圖譜 、聊天機器人
外文關鍵詞： Interpersonal relations extraction, Knowledge graph, Chatbot
相關次數： 點閱：96  下載：0 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 隨著網路的進步，人們的生活模式也逐漸開始改變，購物、社交和閱讀新聞等日常生活都可以在網路上完成。
    然而新聞業提供網路新聞的方式和營運的邏輯卻沒有太大的改變，新聞業者提供報導，而閱讀者只能被動的接受新聞，如同舊時代的紙本報章雜誌，無法與之互動。
    在新聞媒體電子化的現代，我們希望可以提供閱聽者一個可以和新知互動的機會，並且有經過彙整和格式化的圖表可以讓新聞變得一目了然，為此，我們提出一個新聞聊天機器人和知識圖服務，專門解析和名人有關的資訊並給予使用者回饋，希望可以給使用者更好的新聞服務和體驗。
    為了實現上述目標，我們透過網路新聞和維基百科，抽取了名人之間的人際關係和名人的個人基本資料，用知識圖加以儲存，最後根據這些資訊提供使用者恰當的回覆。

    With the rapid evolution of network, many of our daily life can be done online, such as chatting with friends, buying stuff and reading news.
    But the logic and method of news provided by news industry have not changed, they play the role of storyteller, and user can only listen.
    We hope to can carry out graphic and chatting service on celebrity related information to enhance the using Experience in reading web news.
    In order to achieve our goal, we extract interpersonal relationship between celebrity and the basic information of celebrity through online news and Wikipedia. After collecting data, we store them in the form of knowledge graph, and answer user questions accordingly.

    摘要 I Abstract II List of Tables V List of Figures VI Chapter 1 Introduction 1 1.1 Background 1 1.2 Motivation 2 1.3 Goal 3 1.4 Method 4 1.5 Contribution 5 Chapter 2 Related Work 6 2.1 Relation Extraction 6 2.2 News Application 6 2.3 Chatting Service 7 Chapter 3 Method 8 3.1 System Architecture 8 3.2 Preparation 9 3.2.1 Data Collection 10 3.2.2 Celebrity Attribute Extraction 11 3.2.3 Constructing Relationship Lexicons using E-Hownet 12 3.2.4 Relation Pattern Labeling 17 3.3 Constructing Knowledge Graph 19 3.3.1 Celebrity Entity Recognition 19 3.3.2 Interpersonal Relation Extraction 20 3.3.2 Knowledge Graph Construction 22 3.4 Chatting 22 3.4.1 Intent Classification 23 3.4.2 Service Matching 23 3.4.3 Response Generation 24 Chapter 4 Experiments 26 4.1 Dataset 26 4.2 Evaluation Metrics 27 4.3 Experiment on Celebrity Name Verification 28 4.4 Experiment on Intent Classification 30 Chapter 5 Conclusions 32 5.1 Future Work 32 Reference 33

    [1] “Google Trends” [Online]. Available: https://trends.google.com/
    [2] Yuen-Hsien Tseng, Lung-Hao Lee, Shu-Yen Lin, Bo-Shun Liao, Mei-Jun Liu, Hsin-Hsi Chen, Oren Etzioni and Anthony Fader. “Chinese Open Relation Extraction for Knowledge Acquisition” 2014. ACL.
    [3] Zhaohui Wu, Chen Liang and C. Lee Giles. “Storybase: Towards Building a Knowledge Base for News Events” 2015. AFNLP.
    [4] Yu-Hsuan Cheng. “News Bot: A Conversational News Service Based On Event-centric Knowledge Graphs and E-HowNet” 2017
    [5] Takahiko Ito, Shintaro Inuzuka, Yoshiaki Yamada and Jun Harashima. “Real World Voice Assistant System for Cooking” 2019. ACL
    [6] “udn.com” [Online]. Available: https://udn.com/news/index
    [7] “Ckip Tagger” [Online]. Available: https://github.com/ckiplab/ckiptagger
    [8] “Wikipedia” [Online]. Available: https://www.wikipedia.org/
    [9] “E-HowNet” [Online]. Available: http://ehownet.iis.sinica.edu.tw/index.php.
    [10] “D3.js” [Online]. Available: https://d3js.org/

    下載圖示 校內：2025-08-20公開
    校外：2025-08-20公開
    QR CODE