電腦輔助記憶系統之研究與製作∼諧音研究與文章斷詞分析

人的一生中，總是離不開數字，舉凡生日、身份證字號、信用卡卡號、電話號碼、提款卡卡號、密碼等，都與數字息息相關。面對林林總總的數字，除非常用，否則極少部分的人能將其全部數字完全記憶起來。市面上有許多廣告與流行術語，利用諧音來針對無意義、抽象的事物做記憶聯想，將數字轉換成中文諧音字來使得數字醒目又好記。如：000919（鈴鈴鈴，救一救）等。在本論文中，我們研究如何輔助人類方便且容易記憶其相關數字，將一連串無意義的數字，透過電腦輔助，轉換成有意義且容易記憶的詞語。使得人們在面對一連串的數字時，能夠很快地將數字做一中文詞句的聯結以達到易記的目的。研究中針對國語音節聲韻分析、諧音規則、中文斷詞、自然語言詞彙、及構詞組句基本剖析。作一深入的研究與分析，並設計出一套電腦輔助記憶系統。系統對於「無意義的數字轉換成中文助憶詞句」提供一條記憶的捷徑。實驗評估結果顯示：在數字與中文助憶碼的選擇上，有七成多的使用者願意選擇以中文助憶碼來幫助記憶數字；在輔助記憶成效上，經短期記憶與長期記憶的實驗分析，均可達到不錯的助憶效果。

關鍵字

助憶；中文助憶碼；電腦輔助記憶系統；諧音；近似音；國語音節；中文斷詞；漢語構詞

並列摘要

Human beings are closely related to numbers throughout their lifetime. For example, the birth date, identification number, credit card number, telephone number, ATM number, password, etc. all have something to do with numbers. Among plenty of numbers, except for those frequently used, very few people can remember the numbers by heart. In the commercial market, there are many advertisements and popular jargons using homonyms to associate the meaningless and abstract events and further turning numbers into homonyms to make numbers distinct and easy to be remembered. In this thesis, we study how to assist people to remember quickly and easily a series of meaningless numbers by changing them into meaningful and mnemonic Chinese terms through the assistance of computers. In a word, the purpose of this thesis is to associate a series of numbers with mnemonic Chinese phrase or sentence immediately when people encounter those numbers unexpectedly. In the research, we analyze the characteristic of Mandarin syllable, rule of homonyms, Chinese word segmentation, Chinese vocabulary, Chinese word-composition, etc. After a deep research and analysis, we have designed a Computer-Assisted Mnemonic System. This system can offer a shortcut to our memory concerning “meaningless numbers transformed into Chinese Mnemonic words”. According to the experimental results, 70 percent of the testers were willing to use the Chinese mnemonic words and found that it has reached the goal of helping people remember.

並列關鍵字

mnemonic ； Chinese mnemonic words ； computer-assisted mnemonic system ； homonyms ； similar syllable ； characteristic of Mandarin syllable ； Chinese word segmentation ； Chinese word-composition

參考文獻

【1】 K.J. Chen and S.H. Kiu,1992, “Word identification for Mandarin Chinese sentences”, Fifth International Conference on Computational Linguistics,pp.101-107。

【21】陳鳳儀、蔡碧芳、陳克健、黃居仁(1999)：中文句結構樹資料庫的構建。中文計算語言學期刊，88年8月，頁87-104。

【2】 R. Sproat and C. Shih,1990, “A statistical method for finding word boundaries in Chinese text”, Computer Processing of Chinese and Oriental Languages, pp.336-351。