政大機構典藏-National Chengchi University Institutional Repository(NCCUR):Item 140.119/55433
English  |  正體中文  |  简体中文  |  Post-Print筆數 : 27 |  Items with full text/Total items : 94890/125432 (76%)
Visitors : 30648271      Online Users : 281
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
    Please use this identifier to cite or link to this item: http://nccur.lib.nccu.edu.tw/handle/140.119/55433

    Title: 機率式建模技術與自然語言的標記、認知和教學 (I)
    Other Titles: Probability-Based Techniques for Model Construction and Tagging, Cogintion, and Education of Natural Languages
    Authors: 劉昭麟;蔡介立;高照明
    Contributors: 國立政治大學資訊科學系
    Keywords: 機率式建模技術;自然語言;標記;認知;教學
    Date: 2008
    Issue Date: 2012-11-12 11:03:02 (UTC+8)
    Abstract: 資訊科技除了可以協助探索生物的基因和人體醫療的相關資訊之外,是否可以用來協助我們探索人類的心理狀態和認知歷程?這一個研究方向並非今天才有學者提出來,但是這樣的研究議題,在近年來確實才逐漸在學術圈內受到更多的重視。過去三年,我們已經累積了利用貝氏網路和相關的機器學習技術,透過一些學生外顯的表現,來猜測學生學習複雜觀念的學習歷程的研發經驗。這一個實驗室同時也累積了許多自然語言處理的經驗,在中文訴訟文書的處理和中英文電腦輔助教學環境兩個方面都已經建構了實用系統的雛形。基於我們所累積的研發經驗,加上我們所觀察的研究趨勢,我們提出這一個融合人工智慧和認知科學的研究方向,除了學理的研究之外,我們希望能夠把抽象的理論應用在具體的語言教學上。具體地說,我們計畫延伸現有關於貝氏網路等以機率理論作為基礎的機器學習技術,建立一個讓我們可以用比較有效率的方式應用機器學習技術來建構模型的軟體環境。我們計畫利用政治大學所購置的眼動儀,研究不同背景的受試者如何透過眼睛來閱讀文字資訊,藉此我們不僅可以瞭解中文使用者的閱讀歷程,也可以應用研究所得的知識,來檢驗語言學中的計算語言學或者資訊科學中的自然語言處理的各種技術和理念的合理性。透過與政治大學心理系蔡介立教授和台灣大學外語系高照明教授的通力合作,我們相信這個研究計畫不僅在學理上具有重大意義,而且也有很好的機會改進電腦輔助語文教學的實務應用。
    Can we apply computational methods to help the study of human mind, after researchers have shown that computers are helpful for bioinformatics and medical informatics? In fact, applying computational methods to assist us to learn about human mind is not a wild imagination, and had been proposed a long while ago. It is just that this research topic has been receiving more and more attention in recent years, partially due to the tremendous improvement in computational powers of modern computers and partially due to the success achieved in bioinformatics. In the past few years, we have applied probabilistic methods and other machine learning techniques to study the learning process of how students learn complex concepts. We assumed the availability of students’ responses to test items, and attempted to find the best model of learning process based on students’ item responses. We have also applied techniques for natural language processing (NLP) to process judicial documents in Chinese, and have applied NLP techniques to facilitate the preparation of test items for learners and teachers for English and Chinese. At the time of writing, we have implemented usable prototypes for legal informatics and for computer-assisted item writing. Based on the research experience that we gathered in the past years and based on our observation about the trend of research and about the needs in realistic applications, we propose this research plan which attempts to integrate the research work in artificial intelligence and cognitive science. We hope and believe that we can contribute not only to computer and cognitive sciences but also to their applications in language learning. We would like to achieve multiple goals in this three-year project. For the research on computational methods, we will extend our current study on Bayesian networks, probabilistic reasoning, and other relevant machine learning methods. We will also integrate as many machine learning techniques, including those that we will and have developed, in an environment so that people can build models of interest in a more efficient way. For the research on cognitive science, we will employ the eye tracker, which will be offered by the laboratory led by Professor Tsai of National Chengchi University, to study how human subjects of different backgrounds process Chinese text with their eyes. The understanding of how human subjects process text is a very interesting topic itself; it also sheds light on the rationale of the techniques discussed in computational linguistics and natural language processing. The main participants of this research project include Chao-Lin Liu of the Department of Computer Science, Professor Jie-Li Tsai of the Department of Psychology of the National Chengchi University, and Professor Zhao-Ming Gao of the Department of Foreign Languages and Literatures of the National Taiwan University. As a team, we have covered the domain knowledge in computer science, cognitive science, and computational linguistics. We believe that we are prepared to execute this research work in a good manner, and produce appropriate results in the years to come.
    In the first half of this project, we continued what we have been doing in the past few years. We worked on the construction of student models using a probability-based approach, and continued to publish research papers. We have also applied the techniques for natural language processing to computer assisted language learning. In the past several months, we have focused on research issues regarding incorrect Chinese words and regarding the reconstruction of scrambled sentences. In addition, in order to offer better assistance in learning languages, we worked with Professor Tsai of the Department of Psychology (National Chengchi University) to study how native speakers of Chinese move their eyes while they read Chinese. Overall, we published 11 papers in the past 10 months. Two of them are journal articles, three are international conference papers, and six are domestic conference papers. Two other submitted papers are still under review.
    Relation: 基礎研究
    研究期間:9708~ 9807
    Data Type: report
    Appears in Collections:[Department of Computer Science ] NSC Projects

    Files in This Item:

    File Description SizeFormat
    972221E004007MY2(第1年).pdf581KbAdobe PDF347View/Open

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

    社群 sharing

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