臺灣博碩士論文加值系統

摘要
網路蓬勃發展，形形色色的資訊充斥整個全球網路，許多人得以在網路上獲取資訊。網路上的資訊幾乎涵蓋了所有的領域，對資訊需求者而言，無疑是一大知識的寶庫。
如何在極大量的資訊中，快速且準確地找到使用者所需要的資訊，為目前迫切解決的問題。為解決此問題，必須重視使用者個人不同的資訊需求與偏好，則適性化資訊檢索為必然之趨勢。適性化資訊檢索即是：依使用者的不同資訊需求與偏好，提供不同的搜尋策略，藉以提高檢索品質與效率，滿足個人的需求。適性化的目標有三點：(1) 認知個別使用者不同資訊需求，(2) 對使用者偏好的改變，做辨識與適性，(3) 調整使用者認知與資訊內容的差異，以提供較有效較正確的資訊索引。
然而本研究調查發現，目前網路上著名的搜尋引擎並沒有適性化的功能，而有關適性化資訊檢索的文獻也是寥寥無幾。我們認為要達到適性化之目標，則必須針對使用者語意偏好著手，因為使用者的不同知識或文化背景、資訊偏好、表達之查詢概念，都會經由查詢語句與檢索結果的選擇表現出來，所以要瞭解使用者偏好及需求做到適性化，檢索系統必須分析使用者輸入的查詢及結果選擇之間的關係，藉以學習使用者的語意概念、對字義的認知。
一般資訊分類的普遍，已將概念近似之資訊歸為一類，而資訊檢索的定義為：為使用者找出概念上相同或近似的資訊。所以若架構在資訊已正確分類的前題下，檢索系統僅需要找出相關的資訊類別，即可找到相關資訊。由於目前許多資訊都是由人工分類的，其正確性相當高。瓶頸即在將使用者的查詢概念對映到資訊類別，為了同時達到適性化之目標，我們提出適性化資訊檢索架構，是結合案例式推理與機器學習的技術來實現。我們將資訊類別視為一個案例，是一種資訊需求，配合機器學習的技術來分析使用者偏好語意，學習使用者對字義的認知，建立使用者概念與資訊類別間的索引。
為了能正確的瞭解使用者概念，我們設計訣竅式語意模型來 (Heuristic Semantic Pattern) 學習使用者語意偏好。此訣竅式語意模型是架構在一案例樹上，依資訊類別間包含與從屬關係，來分析關鍵字的概念，並用以來學習使用者的語意偏好。
本文將以實驗證明訣竅式語意模組的效益。實驗結果顯示本文提出的訣竅式語意模型在精確率與檢出率，都能有明顯的改善，同時提昇資訊檢索的效率與品質。而評估訣竅式語意模型已能達成適性化的三個目標。

Abstract
With the explosive growth of on-line information on various kinds of platforms (e.g. the library and the Internet), users may have more opportunities to get information conveniently. However, without an efficient and personalized information retrieval (IR) system, the users could also be misleader in the huge amount of information.
Therefore, how to search for useful information for each individual users has become an urgent problem for most researchers in IR, library science and the Internet. To attack the problem, IR systems should be able to consider different users' information needs. In this thesis, I propose a model for building adaptive information retrieval (AIR) systems which may adapt its search strategies to users' different information needs and preferences. Thus, both the quality and the efficiency of IR may be promoted.
In particular, AIR should be able to (1) recognize the personal information needs of individual users, (2) detect the change of user preferences, and (3) map the user’s semantic preferences to the contents of information. However, according to the survey conducted in the thesis, all search engines on the Internet are unable to achieve the three tasks.
Because users' information needs and preferences are often expressed in their queries, AIR should observe and acquire users' semantic preferences from their queries. The output of the learning module is the mapping between query terms and their suitable subset of the document database. As next query is entered, its terms are extracted and mapped to suitable document database in which useful information is more likely to be found.
AIR employs a set of the Heuristic Semantic Patterns (HSPs) to learn users' semantic preferences. The HSPs works on a tree-structured document database which is common for most libraries and web sites on the Internet. A semantic preference of a term is expressed as a mapping between the term and its suitable document category on the document database. As more semantic preferences of query terms may be acquired, AIR may adapt its search strategy to individual needs and preferences.
In this thesis, AIR is explored both theoretically and empirically. The impacts and contributions of the work will be evaluated in terms of the extent to which both the quality and the efficiency of the IR are improved.

目 錄
誌 謝‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧I
中 文 摘 要‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧II
英 文 摘 要 (Abstract) ‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧IV
目 錄‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧VI
圖 目 錄‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧X
表 目 錄‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧XI
第一章
1-1研究主題‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧ 2
1-2研究動機‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧ 3
1-3研究方式與步驟‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧ 4
1-4本文結構‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧ 4
第二章
2-1 現況調查‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧5
2-2 調查結果分析‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧7
第三章
3-1 資訊檢索基本技術‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧9
3-1-1 文件表達方式與結構‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧10
3-1-2 關鍵詞擷取與比對‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧11
3-1-3 向量空間模型 (Vector Space Mode, VSM) ‧‧‧‧‧‧‧11
3-1-4 隱藏語意索引 (Latent Semantic Index, LSI) ‧‧‧‧‧‧12
3-1-5 查詢擴充 (Query Expansion) ‧‧‧‧‧‧‧‧‧‧‧‧12
3-2 智慧型資訊檢索‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧13
3-2-1 依應用領域區分‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧16
3-2-2 依技術領域區分‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧17
3-2-3 依輸入格式區分‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧17
3-2-4 分析與探討‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧18
第四章
4-1 適性化資訊檢索架構 (Adaptive Information Retrieval, AIR) ‧22
4-2 訣竅式語意模型 (Heuristic Semantic Pattern, HSP) ‧‧‧‧‧24
4-3 實作探討‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧26
4-3-1 案例樹 (Case Tree) 建立與初值化方法‧‧‧‧‧‧‧‧26
4-3-2 學習 (Learning) ‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧27
4-3-3 字庫 (Dictionary) ‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧29
4-3-4 關鍵字擷取 (Keyword Extraction) ‧‧‧‧‧‧‧‧‧‧30
4-4 實例說明‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧31
第五章
5-1 實驗方法‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧35
5-1-1 目的‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧35
5-1-2 學習階段 (Training Phase) ‧‧‧‧‧‧‧‧‧‧‧‧‧35
5-1-3 測試階段 (Testing Phase) ‧‧‧‧‧‧‧‧‧‧‧‧‧35
5-1-4 預設關鍵字權重不變‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧36
5-1-5 Cases與Links層次‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧36
5-1-6 評量標準‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧36
5-2 模擬傳統資訊檢索系統‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧37
5-3 實驗資料‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧38
5-4 前置作業‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧38
5-4-1 實驗樣本‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧38
5-4-2 字庫‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧39
5-4-3 資料庫‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧39
5-5 實驗結果分析‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧40
5-5-1 以Case為單位的結果‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧40
5-5-2 以Link為單位的結果‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧44
5-5-3 第二訣竅式語意模型之效能分析‧‧‧‧‧‧‧‧‧‧47
5-5-4 實驗結論‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧48
第六章
6-1 適性化之目標‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧50
6-1-1 AIR之適性化效益‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧51
6-1-2 相關文獻探討‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧53
6-2 資訊檢索之品質‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧54
6-3 資訊檢索之效率‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧56
6-4 未來展望‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧56
6-4-1 其他訣竅式語意模型之設計與應用‧‧‧‧‧‧‧‧‧56
6-4-2 AIR應用探討‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧58
第七章
7-1 資訊檢索現況‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧60
7-2 適性化資訊檢索‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧60
7-3 適性化資訊檢索之實用價值‧‧‧‧‧‧‧‧‧‧‧‧‧62
參考文獻‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧64
附錄 A 案例樹 (Case Tree) ‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧69
附錄 B 實驗查詢語句‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧‧72

參考文獻
[Allan ‘96] Allan J., ”Incremental Relevance Feedback for Information Filtering,” Proc. of SIGIR 1996.
[AltaVista ’99] http：//altavista.digital.com/, 1999 March.
[Appleton ‘92] Appleton E. L., “Smart Document Retrieval” Journal of DATAMATION, Vol.38, No.2, 1992.
[Armstrong et al. ‘95] Armstrong R., Freitag D., Joachims T., and Mitchell T., ”WebWatcher: A Learning Apprentice for the World Wide Web,” AAAI Spring Symposium on Information Gathering from Heterogeneous. Distributed Environment. March. 1995.
[Balabanovic & Shoham ] Balabanovic M. and Shoham Y., “Learning Information Retrieval Agents: Experiments with Automated Web Browsing,”
[Belkin & Croft ’92] Belkin N. J. and Croft W. B., “Information Filtering and Information Retrieval: Two Sides of the Same Coin?,” Communication of the ACM, December 1992, Vol. 35, No. 12, pp29-38
[Bloedorn et al. ‘96] Bloedorn E., Mani I., MacMillan T. R., “Machine Learning of User Profiles： Representational Issues,” Proc. of AAAI 1996.
[Burke et al. ‘97] Burke R. D., Hammond K. J., & Kulyukin V. J., “Question Answering from Frequently-Asked Question Files: Experiences with the FAQ Finder System,” The University of Chicago Computer Science Department, Technical Report TR-97-05, June 1997.
[Chien ‘95] Chien Lee-Feng, “Fast and Quasi-Natural Language Search for Gigabytes of Chinese Texts,” SIGIR 1995.
[Daniels & Rissland ‘95] Daniels J. J., and Rissland E. L., “Case-Based Approach to Intelligent Information Retrieval,” Proc. of SIGIR 1995.
[Domeshek et al. ‘96] Domeshek E., Kedar S., and Gordon A., “Interactive Information Retrieval Systems with Minimalist Represnetation,” Proc. of AAAI 1996.
[Faloutsos & Oard ‘96] Faloutsos C. and Oard D., ”A Survey of Information Retrieval and Filtering Method”, http://www.cs.jhu.edu/~weiss/paper2.html, 1996.
[Filman & Pant ’98] Filman R. E. and Pant S., “Searching the Internet,” IEEE Internet Computing, July. August 1998.
[Fitzpatrick & Dent ‘97] Fitzpatrick L. and Dent M., ”Automatic Feedback Using Past Queries: Social Searching?,” Proc. of SIGIR. 1997.
[GAIS ‘99] http：//gais.cs.ccu.edu.tw/cwww2.html, 1999 March.
[Go2Net ‘99] http://www.go2net.com/index.html, 1999 March.
[Hsinchun et al. ‘98] Hsinchun C., Chung Y-M., Ramsey M., and Yang C. C., “An intelligent personal spider (agent) for dynamic Internet/Intrant searching,” Decision Support Systems 23, pp41-58, 1998.
[Kristensen ‘93] Kristensen J., “Expanding End-Users’ Query Statements for Free Text Searching with A Search-Aid Thesaurus,” Information Processing Management Vol. 29, No. 6, pp. 733-744, 1993.
[Lam et al. ‘96] Lam W., Mukhopadhyay S., Mostafa J., and Palakal M., ”Detection of Shifts in User Interests for Personalized Information Filtering,” Proc. of SIGIR 1996.
[Lassila ‘98] Ora Lassila, “Web Metadata: A Matter of Semantics,” IEEE Internet Computing, July. August 1998.
[Lawrence & Giles ‘98] Lawrence S. and Giles C. L., “Context and Page Analysis for Improved Web Search,” IEEE Internet Computing, July. August 1998.
[Lewis ‘96] Lewis D. D., “Natural Language Processing for Information Retrieval,” Communication of ACM, January 1996, Vol. 38, No. 1.
[Li ‘98] Li Y., “Toward A Qualitative Search Engine,” IEEE Internet Computing, July. August 1998.
[Liu R-L & Chien J. ‘99] Rey-Long Liu and John Chien, “A Self-Extensible Hybrid Reasoning Mode and Its Application to Adaptive Information Retrieval,” submitted to Knowledge & Information System: An International Journal, revised. 1999.
[Lycos ’99] http：//www.lycos.com/, 1999 March.
[InfoSeek ’99] http：//www.infoseek.com/, 1999 March.
[Khan & Card ‘98] Khan I. and Card H. C., ” Adaptive information agents using competitive learning,” Journal of Network and Computer Applications n.21000 pt.2 pp.69-90, 1998.
[Maes ‘94] Maes P., ”Agent that Reduce Work and Information Overload,” Journal of CACM, July, Vol.37,No.7, 1994.
[McCune et al. ‘85] McCune B. P., Tong R. M., Dean J. S., and Shapiro D. G., “RUBRIC: A System for Rule-Based Information Retrieval,” IEEE Transactions on Software Engineering, VOL. SE-11, NO. 9, September 1985.
[MetaCrawler ’99] http：//metacrawler.cs.washington.edu/, 1999 March.
[Monster ’99] http：//www.monster.com.tw/, 1999 March.
[Rajaraman & Norvig ‘98] Rajaraman A. & Norvig P., “Virtual Database Technology: Transforming the Internet into a Database,” IEEE Internet Computing, July. August 1998.
[Rissland et al. ‘94] Rissland E. L., Skalak D. B., and Friedman M. T., “Heuristic Harvesting of Information for Case-Based Argument,” Proc. of AAAI, 1994.
[Stein ‘91] Stein R. M., “Browsing Through Terabytes,” Journal of BYTE, Vol.16, No.5, 1991.
[Sycara & Miyashita ‘94] Sycara K. and Miyashita K., “Case-based Acquisition of User Preferences for Solution Improvement in Ill-Structured Domains,” Proc. of AAAI, 1994.
[Tseng et al. ‘94] Tseng C-M., Lang C-Y., and Chien L-F., “An Approach to Extending Record-based Database Management System for Chinese Full-text Indexing and Searching” The 5th International Conference on Information Management 1994國際資訊管理研討會.
[Vollrath et al. ‘98] Vollrath I., Wilke W., and Bergmann R., “Case-based Reasoning Support for Online Catalog Sales,” IEEE Internet Computing, July. August 1998.
[Weiss et al. ‘96] Weiss S. A., Kasif S., and Brill E., “Toward An Adaptive Framework for Information Retrieval,” AAAI Spring Symposium on Information Retrieval 1996.
[WhatSite ‘99] http：//www.whatsite.com/, 1999 March.
[Yahoo ’99] http：//chinese.yahoo.com/, 1999 March.
[YAM ’99] http：//search.yam.org.tw/b5/search/, 1999 March.
[中華民國期刊論文 ‘99] 中華民國期刊論文索引影像系統，http://readopac.ncl.edu.tw/cgi/ncl3/m_ncl3 1999 March.
[史孟蓉 & 劉瑞瓏 ‘98] 史孟蓉、劉瑞瓏, "適性化主管資訊代理人," 第四屆國際資訊管理研究暨實務研討會論文集, 臺北. 1998
[吳忻萍 ‘97] 吳忻萍 “以隱藏語意索引為基礎之中文資訊檢索” 國立台灣大學資訊管理研究所碩士論文, 1997
[陳秀美 ‘96] 陳秀美, “ WWW 檢索工具的選擇與比較,” 資訊傳播與圖書館學 第二卷 第四期, 1996.
[黃雲龍 ‘97] 黃雲龍 “中文全文文件群集索引理論研究--向量空間模型的建構” 國立台灣大學商學研究所碩士論文, 1997
[黃森青 ‘96] 黃森青, “Automatic Classification of Chinese Documents,” 國立交通大學資訊科學研究所碩士論文, 1996
[曾元顯 ‘97] 曾元顯, “關鍵詞自動擷取技術與相關詞回饋,” 中國圖書館學會會報59期, Nov. 4, 1997.
[博譽台灣法律資料庫 ‘99] 博譽台灣法律資料庫，http://www.law.com.tw/ 1999 March
[壽大衛 ‘98] 壽大衛, “智慧型全球資訊網資訊考掘,” ICCAI 1998.
[簡立峰 ‘96] 簡立峰, “尋易系統(Csmart)與中文智慧型資訊檢索,” 21世紀資訊科學與技術的展望國際學術研討會, 1996
[簡鳳江 & 劉瑞瓏 ‘98] 簡鳳江、劉瑞瓏，”適性化網路資訊檢索之現況與展望”，跨世紀資訊科技研討會論文集，1998
[顧皓光 ‘97] 顧皓光 “網路文件自動分類” 國立台灣大學資訊管理研究所碩士論文, 1997