中文繁簡在字體或電腦編碼上明顯不同之外，在部份詞彙的用法也有不同，這些用法不同的詞彙卻有相同意義的詞彙稱為繁簡體中的等義詞，這些等義詞在雙方文化交流時可能會造成一些障礙，例如人們互相對話、文件書籍翻譯或軟體系統等轉換時容易造成詞義上的誤解，目前均以人工方式來解決不同詞彙的問題，均會費時耗力且易疏漏，若能利用科學的方法讓電腦能自動辨識中文繁簡的等義詞，便能利用辨識出的等義詞給予提示，解決繁簡詞義不同所造成的誤解。 依照實驗設計架構，首先建立電腦類與一般類的繁簡體語料庫，作為辨識的基礎，並建立研究的架構與方法，分為二個階段三種方法，第一階段使用第一種方法，我們先使用N-gram辨識等義詞，評估單一方法是否能有效辨識出等義詞，第二階段使用第二種方法PMI-IR & LC-IR方法與第三種方法Context Vector，評估第二階段的方法是否能將等義詞的辨識能力提高。 根據本研究目的，讓電腦能自動在語料庫中自動辨識中文繁簡等義詞，所以提出了新的辨識架構，用N-gram初步辨識出等義詞，並經由PMI-IR & LC-IR與Context Vector方法提高Precision約0~20%不等。本研究結論是採用不同語言的語料庫，使用N-gram能夠辦識出等義詞，並搭配PMI-IR & LC-IR與Context Vector方法後，可以強化與提昇其等義詞辨識的能力，解決單一方法等義詞辨識能力不足之問題。 / Traditional Chinese and Simplied Chinese are not only different in the typeface and in the computer code, but also in the partial usage of vocabularies. These vocabularies which have different usage but have the same significance are called synonyms. These synonyms will cause some obstacles and misunderstanding in meaning when two parties have cultural exchange, such as during conversation, documents and books translation or softwares system transformation. What we do to solve the problem now is picked them out by manpower, but that will waste a lot of time and strength and easily make errors. If we can use scientific way to let the computer distinguish automatically the synonyms between Traditional Chinese and Simplied Chinese, we will be able to solve such misunderstanding by the hints of the distinguished synonyms. According to the structure of experiment, to let the computer distinguish automatically the synonyms between Traditional Chinese and Simplied Chinese, we have to establish a Traditional Chinese and Simplied Chinese computer category and a general category first as the basis of identification. We should build up the research structure and the method, which divided into two stages and three methods. The first stage uses the first method to use N-gram to distinguish the synonyms and then review if this single method can identify the synonyms effectively. The second stage uses the second method PMI-IR & LC-IR and the third method Context Vector and review if the second stage can raise the synonyms’ ability of identification. According to this research purpose, the computer to study on automatic exact recognition synonyms between traditional and simplified Chinese, so has proposed the new structure of distinguishing, N-gram automatic exact recognition synonym tentatively, and PMI-IR & LC-IR and Context Vector method can improve Precision about 0~20%. This conclusion is a corpus base of using different languages, using N-gram can be exact recognition synonyms, PMI-IR & LC-IR and Context Vector method, can improve single method ability.

本篇論文的主旨在研究「也」「又」「還」三個連接詞在漢語敘述文中的 篇章功能。過去對於這三個連接詞的研究，多半偏向句法和語意關係的探 討，而事實上，我們還可以從另一個角度來分析這三個連接詞，那就是探 討它們的篇章功能。研究這三個連接詞的篇章功能之前，我們首先根據 Matthiessen 和 Thompson 的模式分析了漢語句群間的凝聚關係(text relations among combined clauses)。 此一分析不僅進一步證明「也」 「又」「還」確實可以作為連接詞使用，同時更清楚地呈現出「也」「又 」「還」所連接的子句之間的關係，以及這些子句本身與其前後子句所形 成的凝聚關係。「也」「又」「還」作連接詞用時，是用來連接一組語意 相關的子句。它們都可以用來表示聯合關係；不過，「也」和「又」還可 以用來表示對照關係。表示聯合關係時，「也」連接二個或二個以上平列 的子句；「又」也是連接平列的子句，然而「又」進一步加強這些子句所 呈現的聯合關係；「還」連接帶有輕重之別的子句。至於如何知道「又」 所加強的是什麼，以及如何決定那個子句比較重，這都得依上下文而定。 表示對照關係時，「也」和「又」都是用來連接二個看似相反，而實質上 卻相輔相成的子句，差別是「又」所連接的二個子句呈現出說話者矛盾的 心理狀態。上文所提出的假設，經由我們分析書寫的語料而獲得驗證。除 此之外，我們還做了一項問卷調查，目的在測試中國人如何使用這三個連 接詞，這項問卷的結果顯示我們的假設大致正確。最後，我們依據語料的 分析及問卷結果的討論，將上文所提的假設做了小幅度的修正，使其更能 明確地說明這三個連接詞的篇章功能。

近年來，基因與疾病之關聯分析 (association analysis) 越來越受到研究學者重視，因為在複雜性疾病與易感性基因之探討中 傳統的連鎖方法 (linkage method) 已不適用，所以複雜性疾病與易感性基因的關聯分析也蓬勃發展起來。在本文中我們主要是在探討 關聯分析中以家庭為研究資料與以群體為研究資料之間的優缺點，進而取長補短提出結合兩種資料之新的關聯分析方法 來增加估計與檢定之效力。我們同時考慮環境因素，探討基因因素與環境因素之交互作用。 本研究共分為三部份。第一部份探討如何整合病例-父母/病例-同胞 (case-parent/case-sibling) 與病例-對照 (case-control) 研究。我們提出一個加權最小平方 (Weighted Least Squares) 的方法將病例-父母/病例-同胞與病例-對照分析之估計式加以結合，以增進統計檢定之效力。 第二部分旨在探討基因-環境之交互作用。我們提出一個二階段研究設計法。在第一階段研究中，先收集病例資料; 在第二階段研究中，再收集其相對應之控制組資料。我們提出一個迴歸估計式以結合第一階段之單純病例分析(case-only analysis) 與第二階段之病例-對照分析。此建議之估計式即使在基因因子與環境因子 獨立之條件 (此條件為單純病例分析所必需) 不成立的情形下，依然可得出正確之統計推論。 第三部份旨在探討群體分層 (population stratification) 存在 之情形下，基因-環境之交互作用。我們提出一個二階段研究設計，以病例資料為第一階段資料， 再從病例資料中隨機抽取一部份病例患者之父母資料為第二階段資料。我們提出一個迴歸估計式結合單純病例研分析與病例-父母分析之估計式。 此新估計式即可整合單純病例分析與病例-父母分析，同時在群體分層存在之情形下，仍可得出有效之統計推論。 / In recent years, there are increasing attention to association studies, because linkage method will not be suitable under complex disease and susceptible genes. In the thesis, we are probing into association of family study and population study. And we combine family study and population study for increased efficiency of association method. We also consider interesting studies about gene-environment interactions. The thesis contains three projects. The first project focuses on examining when and how the two sources of information offered by such studies, one from the case-parent/case-sibling analysis, and the other from the case-control analysis with data from affected subjects and unrelated controls, can be integrated to enhance statistical power. We propose a weighted least-squares approach to linearly and optimally combine separate estimators from the case-parent/case-sibling and the logistic regression analysis for the association parameters. In the second project, we focus on examining the situation of gene-environment interaction. We propose a two-stage design. In the first stage, we collect patient data, and we seek out control data with respect to cases in the second stage. We propose regression analysis estimation in order to combine the case-only analysis in the first stage and the case-control analysis in the second stage. This estimation earns the correct statistical inference when genes and environment factors are not independent. In the third project, we explore gene-environment interactions under population stratification. We propose a two-stage design. In the first stage, we collect patient data, and we randomly collect a partial data of patient's parent from the cases in the second stage. We propose regression analysis estimation in order to combine the case-only analysis and the case-parent analysis. This estimation can combine the case-only analysis and the case-parent analysis, and attains effective statistical inference under population stratification.

在許多實際情錶里, 例如工業、農業及生物實驗等, 想要在控制一個處理情況下, 同 時比較好幾個試驗處理, 比方說有p 個試驗處理。( 這也是所謂的在控制下, 多重比 較問題。 ) 已從事的研究包括二方向異質性的行列設計(row-column design) 及一個方向上差異 的集區設計(block design)。此篇論文為考慮在集區設計內置行與列(block designs with nested rows and columns)的模型下求得同時以p 組試驗組與一組對照組比較 之A 式最佳設計。所謂A 式最佳設計, 就是在所有設計里, 找出tr(M (d))為最小的 , 因tr(M (d))與 var( )成比例, 所以A 式最佳設計。所謂A 式最佳設 計在統計上之直覺解釋為使α - α 之最佳線性不偏估計式 之變方和為最小 的設計。 A 式最佳設計當p=7,...,25, 集區數b=1,...,50時之表錄詳列於后, 某些特殊形式的 設計亦將在例題中列舉。 本篇論文之目錄如下: 一、摘要 二、文獻回顧 三、引介 四、符號注釋及A 式最佳內置行與列集區設計之條件 五、求最佳內置行與列集區設計之方法 六、特例探討 七、最佳A 式內置行與列集區設計表 八、參考文獻

《古今和歌集》是最早的敕撰和歌集，於延喜五年（905年）成立，其編撰者為紀友則・紀貫之・凡河内躬恒・壬生忠岑。歌數約為一一一〇首，由二十巻構成。在卷別上，分為春（上下巻）・夏・秋（上下巻）・冬・賀・離別・羇旅・物名・恋（巻一～巻五）・哀傷・雑・雑体・大歌所御歌。歌風屬纖細優美。一一一〇首和歌之中，由四季歌（342首）及戀歌（360首）二大支柱構成，四季歌是依據季節的推移來排列。而戀歌是依戀愛的進行過程來排列，其排列被認定具有時間性之推移。因此，本論文將以《古今和歌集》的戀歌為對象，透過「晨與夜」「夢境與現實」「昔時與今日」等時間性之對照表現，來解析戀歌。 本論文首先將和歌分類，整理出同性質之和歌。其次，分析各類和歌中之對照表現，並考察各和歌中對照表現之意義。 本論文由六章構成，第一章論述研究目的等內容。第二章從和歌與戀歌的本質開始，探討《古今集》戀歌的本質及構成，並分析《古今集》戀歌中的時間特質。第三章以「晨與夜」的對照表現為中心。其中分為「特定的晨與夜」、「不特定的晨與夜」二個主題，來探討《古今集》戀歌中晨與夜的對照表現。第四章以「夢境與現實」為中心，其中分為「想於夢中相會」→「無法在夢境相會」→「夢中會面」→「現實中會面」→「無法再於夢境相會」等五個主題來探討。第五章是以「昔時與今日」為中心來分析。分為「相遇之前的昔時與相遇之後的今日」及「相愛的昔時與離別後的今日」二個主題，來考察歌中「昔時」與「今日」的對照表現。第六章是透過與《萬葉集》的比較，來釐清《古今和歌集》中時間性對照表現的角色與意義。 / “Kokinwakashuh” is the first waka collections that Japanese Emperor chose, and it was compiled around the year 905. The editors are Ki no Tsurayuki, Ki no Tomonori, Oshikochi Mitsune, and Mibu no Tadamine. The chapters are Spring, Summer, Fall, Winter, Congratulations, Partings, Travel, Acrostics, Love, Laments, Miscellaneous, Miscellaneous Forms, and Traditional Poems from the Bureau of Song. The style is courtly elegance of classical, disciplined beauty. In these 1100 poems, the four season Poems and ‘koinouta’ (Love Poems) are two pillars. The Seasons Poems are arranged according to the transition of seasons. The ‘koinouta’ (Love Poems) are arranged according to the process of love, and there is the transition of time in the arrangement. It is admitted that time is a critical factor in Poems of “Kokinshuh”. Therefore, the ‘koinouta’ (Love Poems) of “Kokinwakashuh” is the subject in this thesis. The ‘koinouta’ (Love Poems) are analyzed through the contrastive expression of time, like ‘morning and night’, ‘dream and reality’ and ‘past and present’. In this thesis, poems were classified first, and the homogeneous poems were put together. The contrast expression in the poems of each kind was analyzed, and the meaning of the contrast expression in each poem was investigated. This thesis consists of six chapters. The purpose of research was discussed in Chapter 1. In Chapter 2, the theory is progressed from the essence of the waka and the ‘koinouta’ (Love Poems). In the same chapter, essence and the composition of the ‘koinouta’ (Love Poems) were examined, and the time characteristic in ‘koinouta’ (Love Poems) of “Kokinwakashuh” was analyzed. In Chapter 3, it centered on the contrast expression of ‘morning and night’. It is separated for two subjects, ‘specific morning and night’ and ’unspecific morning and night’. In Chapter 4, it centered on the contrast expression of ‘dream and reality’. It is separated for five subjects. In Chapter 5, it centered on the contrast expression of ‘past and present’. It is separated for two subjects. In Chapter 6, the role and the meaning of a time contrast expression in “Kokinwakashuh” were clarified through the comparison with “Manyoshuh”.