運用近似語意相等在模糊資料庫上的模糊查詢

Using Proximity Semantic Equivalence for Fuzzy Query in Fuzzy Databases

黃堯銘(Yao-Ming Huang)

指導教授 ： 劉俞志

元智大學/資訊學院/資訊管理學系/碩士(2002年)

摘要

從Zadeh提出模糊理論至今，已被廣泛的研究與應用到眾多領域上。在模糊理論的眾多應用裡，著眼於處理不明確事務的模糊查詢，能幫助企業組織從大量資料裡，找出接近需求的資訊，協助管理當局的決策制定。
而模糊查詢的研究，從原先以傳統資料庫為背景，在模糊資料庫的發展趨勢下，逐漸地轉移至模糊資料庫領域。再加上模糊領域的發展，使得模糊資料的呈現更趨多元。因此對於各類型的模糊資料之相似程度衡量，通常採用不同的衡量機制。
本論文針對模糊語句在模糊資料庫的查詢，以近似語意相等(PSE)來計算模糊資料間的相似程度。其中查詢與模糊資料可包含語意變數的模糊項、模糊陳述，並容許使用者以修飾詞語和語意法則，對模糊項或模糊陳述加以強調或修飾。本研究在整個模糊查詢程序裡，引用轉換層概念，建構出以單一衡量機制進行各類型模糊資料查詢之架構，並提出一套轉換機制，將各類型的模糊資料轉成分割區歸屬程度表示法，使其可直接計算近似語意相等(PSE)，達到多元化的模糊資料處理功能。
模糊查詢的發展從傳統資料庫到模糊資料庫，增加更多樣的模糊資料展現方式，卻也遞增模糊資料庫系統處理上的麻煩，本論文所提出的模糊查詢架構，能有效地降低模糊資料庫系統處理與維護上的複雜度，讓模糊資料的多樣性與系統效率能同時兼顧，幫助未來在企業組織上的推廣應用與發展。

關鍵字

近似語意相等 ； 模糊查詢 ； 模糊資料庫

並列摘要

ABSTRACT
Fuzzy theory, proposed by Zadeh in 1965 was been researched and applied extensively to several domains. In many applications of fuzzy theory, which focus on handling the search of the uncertain facts can help to find out the most closed requested information, and assist to manage the decision’s making of managers.
In the fuzzy query research, with the development of fuzzy database, has gradually moved its background from traditional database to fuzzy database. And with the development of fuzzy domain, it let the fuzzy data express in diversification, therefore, we usually adopt different measure functions as to measure out the similarity degree of all kinds of fuzzy data.
A thesis advanced in this paper is that for the search of the fuzzy query in the fuzzy database, we use the PSE(Proximity Semantic Equivalence) to compute the similarity of fuzzy data. In there, the query and fuzzy data can include the fuzzy terms and the fuzzy proposition of linguistic variables, and allow the users to use the modifier and semantic rules to emphasis and modify the fuzzy terms also fuzzy propositions. In the whole processes of fuzzy query, we cited the conception of transfer layer as construct a framework of a single measure function as to work with all kind of fuzzy data queries, and brought up a transfer function to transform all kinds of fuzzy data to the same format, let it can compute PSE directly as to approach the diverse fuzzy data handling function.
From traditional database to fuzzy database, it increased more fuzzy data expressions but also gain some problems on the handling of fuzzy database system. The fuzzy query framework proposed in this paper can effective to simplify the complexity on the maintenance of fuzzy database system, let the fuzzy data can take care both its diversification and system efficiency as to apply and spread to the enterprise in the future.

並列關鍵字

Proximity Semantic Equivalence ； Fuzzy Query ； Fuzzy Databases

參考文獻

[1]L.A. Zadeh, "Fuzzy Sets" , Journal of Information and Control, 8, pp.338-353, 1965.

[2]B.P. Buckles and F.E. Petry, "A Fuzzy Representation of Data for Relational Databases" , Fuzzy Sets and Systems, 7, pp.213-226, 1982.

[3]J.F. Baldwin and S.Q. Zhou, "A Fuzzy Relational Inference Language" , Fuzzy Sets and Systems, 14, pp.155-174, 1984.

[4]H. Prade and C. Testemale, "Generalizing Database Relational Algebra for the Treatment of Incomplete/Uncertain Information and Vague Queries" , Information Sciences, 34, pp.115-143, 1984.

[5]E.A. Rundensteiner, L.W. Hawkes, and W. Bandler, "On Nearness Measures in Fuzzy Relational Data Models" , International Journal of Approximate Reasoning, 3, pp.267-298, 1989.