Metasearch engine中資料庫分類之關鍵詞維護機制

Classifier maintenance system of Database classification for Metasearch engine

陳志榮(Zei-Rong Chen)

指導教授 ： 林志麟

元智大學/資訊學院/資訊管理學系/碩士(2002年)

摘要

由於網際網路的資訊成長過於迅速及龐大，搜尋引擎發展至今，尚沒有任一搜尋引擎可完全包含網際網路的所有資訊，導致使用者必須花費額外的時間去尋找最適用的搜尋引擎，而每個搜尋引擎搜尋介面及機制的不同也會造成使用者使用上的困擾。
為解決上述問題，匯總式搜尋引擎(Metasearch engine)在近代被發展來幫助使用者挑選及查詢符合資訊需求的搜尋引擎。匯總式搜尋引擎挑選符合使用者需求的資料庫過程稱為資料庫選擇(Database selection)，而資料庫分類(Database classification)是屬於資料庫選擇過程的一部份，它可提供使用者選擇所要查詢的類別，更有效率地選擇合適的資料庫去執行搜尋動作。資料庫分類須有相當的準確性，若資料庫被分派到不恰當的類別中，則將影響資料庫選擇程序的準確率。
本研究的目的藉由維護類別分類器，避免分類器因時間而趨於老舊(新名詞的出現、舊名詞被取代等)，反覆維護類別分類器可提升分類器的效用，以進行更準確的分類，及預防資料庫性質的變動而影響資料庫選擇程序的效能。

關鍵字

資料庫分類 ； 匯總式搜尋引擎 ； 資訊擷取 ； 關鍵詞擷取 ； 搜尋引擎

並列摘要

The Internet has become a vast information resource in recent years. To help ordinary users find desired data in Internet, many search engines have been created , but no one can contain the all information of Internet. Users are hard to choose which search engines are suitable for use and spend extra time to learn to use each different search engines.
Metasearch engine offer mechanisms to search for the search engine that suitable for user needs. The selection algorithm used by a Metasearch engine to determine whether a search engine should be sent the query. Database classification is part of Database selection process. Database classification have increased selection procedure efficiency. Accuracy is important in Database classification. If Database classification. The wrong classification affect the accuracy and efficiency of database selection procedure.
In this study, A Classifier maintenance system is designed and implemented.
The main propose is to avoid the bad classifyer. The accurate classification procedure could improve efficacy of database selection.

並列關鍵字

Database classification ； Metasearch engine ； Information retrieve ； Keywords retrieve ； Search engine

參考文獻

[2] Search Engine Watch: Tips About Internet Search Engines & Search Engine Submission http://www.searchenginewatch.com

[7] J. Callan, Z. Lu, and W. Croft. Searching Distributed Collections with Inference Networks. ACM SIGIR Conference, 1995.

[8] M. H. Chignell, J. Gwizdka and R. C. Bodner, “Discriminating Meta-Seach: aFramework For Evaluation,” Information Processing and Management, Vol.35, 1999, Pages 337-362.

[9] D. Dreilinger and A. E. Howe, “Experiences with Selecting Search Engines Using Metasearch, ” ACM Transaction on Information Systems, Vol. 15, July1997, Pages 195-222.

[10] W. W. Cohen. Learning trees and rules with set-valued features.