唐代唐玄度的《九經字樣》在唐代正字運動、字樣學、經典文字的研究方面具有重要意義，但由於該書之撰寫動機為補充《五經文字》之不足，並且篇幅不多，故尚未受到人們的重視。因此本文將對唐玄度的《九經字樣》進行全面地探討，其主要內容可舉以下四點： 一、簡要說明唐以前的文字規範情況，並述評唐代正字運動及其影響。 二、釐清《九經字樣》的字數、蒐字範圍、分部與收字歸部、《九經字樣》各 種版本之間的比較等有關體制、版本的問題。 三、探討《九經字樣》所收的各類字，即正字、異體字、通假字之相關問題。 四、以如上的內容為基礎，整理並分析《九經字樣》所樹立的正字標準與正字 觀，及《九經字樣》之優點與缺失、意義和影響。

本研究欲探討我國公司治理相關法規文字及其運作之解讀可能產生之問題，從95 年度證券交易法之修正條文中，辨認以下釋例來檢視問題： 1.「會計主管」與「主辦會計」之意義是否相同？ 2.審計委員會之組成究應包括全部或部分之獨立董事？ 3.獨立董事究應參加一個或多個委員會？ 4.審計委員會及其成員行使職權之方式是否不同？ 5.獨立董事針對證交法第十四之三條之事項所表示之保留意見是否為反對意見？ 此外，本研究另外探討法規閱讀者對以上問題的認知是否因其學業背景而有不同？ 本研究以政治大學法律系、會計系、法律所、會計所、EMBA 學生為受測者，發放問卷，結果發現法規閱讀者對於這些法規用字認知很分歧。立法者在起草這些法規時，應考慮到法規適用者，將法規文字訂定得更加明確，且合理，自然可讀性可提高。 / This research is to discuss about some possible problems in perceiving the law wording in the articles in the Securities and Exchange Act revised in 2006. It examines problems by identifying the following examples: 1. Do ”accounting officer” and “in-charge accountant” mean the same? 2. Shall the audit committees be composed of the entire number of independent directors or just some independent directors? 3. Shall independent directors attend only one or more than one committees? 4. Is there any difference between audit committees and independent directors in the modes of exercise? 5. Does the qualified opinion of independent directors on the Article 14-3 in the Securities and Exchange Act mean dissenting opinion? This research will study if the perceptions of readers differ because of their academic backgrounds. This research uses questionnaires to obtain opinions of students from the Departments of Accounting, Law, and EMBA. The analysis in this research shows that readers have large differences in the perceptions of the laws and regulations. When drafting the laws and regulations, legislators should think about the users. They should make the laws and regulations more definite and reasonable, which will enhance the readability.

面對知識管理時代與學術論文不斷的激增下，快速擷取學術論文的研究主題領域是相當重要的項研究。本研究從文獻結構中分析，應用了特徵擷取、中文字詞分析、統計及演算等方法，對中文學術論文標題及其中文參考文獻標題進行分析，以擷取可描述該篇論文研究主題之專業詞彙。 對應論文的研究主題擷取，本研究建立了專業詞彙資料庫以及虛字符號字庫，應用在研究主題詞彙擷取時，論文標題的詞彙比對。專業詞彙資料庫，包括了各專業詞彙，以及詞彙間的研究主題關係，此外，也包含了詞彙間的同義詞關係。虛字符號字庫，則包含了需要在去除標題中各虛字符號時，所比對的字庫。 在最後的結果分析上，採用詞頻統計統計出論文及其參考文獻標題所出現的專業詞彙次數，並使用論文與參考文獻的對應分析產生六種可能的結果，說明應用此結果加上次數統計的方式，如何確切的擷取可描述該論文研究主題的詞彙。而在研究過程中，本研究以“中文文獻及字詞的資訊化處理”此研究主題的論文為例，建置雛型系統及詞庫來加以探討及印證。

新聞事件可描述為「一個時間區間內、同一主題的相似新聞之集合」，而新聞大多僅是一完整事件的零碎片段，其內容也易受到媒體立場或撰寫角度不同有所差異；除此之外，龐大的新聞量亦使得想要瞭解事件全貌的困難度大增。因此，本研究將利用文字探勘技術群聚相關新聞為事件，以增進新聞所帶來的價值。 分類分群為文字探勘中很常見的步驟，亦是本研究將新聞群聚成事件所運用到的主要方法。最近鄰 (k-nearest neighbor, kNN)搜尋法可視為分類法中最常見的演算法之一，但由於kNN在分類上必須要每篇新聞兩兩比較並排序才得以選出最近鄰，這也產生了kNN在實作上的效能瓶頸。本研究提出了一個「建立距離參考基準點」的方法RTD-based kNN (Relative Text-Distance-based kNN)，透過在向量空間中建立一個基準點，讓所有文件利用與基準點的相對距離建立起遠近的關係，使得在選取前k個最近鄰之前，直接以相對關係篩選出較可能的候選文件，進而選出前k個最近鄰，透過相對距離的概念減少比較次數以改善效率。 本研究於Google News中抽取62個事件(共742篇新聞)，並依其分群結果作為測試與評估依據，以比較RTD-based kNN與kNN新聞事件分群時的績效。實驗結果呈現出RTD-based kNN的基準點以常用字字彙建立較佳，分群後的再合併則有助於改善結果，而在RTD-based kNN與kNN的F-measure並無顯著差距(α=0.05)的情況下，RTD-based kNN的運算時間低於kNN達28.13%。顯示RTD-based kNN能提供新聞事件分群時一個更好的方法。最後，本研究提供一些未來研究之方向。 / News Events can be described as "the aggregation of many similar news that describe the particular incident within a specific timeframe". Most of news article portraits only a part of a passage, and many of the content are bias because of different media standpoint or different viewpoint of reporters; in addition, the massive news source increases complexity of the incident. Therefore, this research paper employs Text Mining Technique to cluster similar news to a events that can value added a news contributed. Classification and Clustering technique is a frequently used in Text Mining, and K-nearest neighbor(kNN) is one of most common algorithms apply in classification. However, kNN requires massive comparison on each individual article, and it becomes the performance bottlenecks of kNN. This research proposed Relative Text-Distance-based kNN(RTD-based kNN), the core concept of this method is establish a Base, a distance reference point, through a Vector Space, all documents can create the distance relationship through the relative distance between itself and base. Through the concept of relative distance, it can decrease the number of comparison and improve the efficiency. This research chooses a sample of 62 events (with total of 742 news articles) from Google News for the test and evaluation. Under the condition of RTD-based kNN and kNN with a no significant difference in F-measure (α=0.05), RTD-based kNN out perform kNN in time decreased by 28.13%. This confirms RTD-based kNN is a better method in clustering news event. At last, this research provides some of the research aspect for the future.

本研究針對半導體產業，提供一智慧型搜尋功能，讓使用者在大量資料中能快速及準確地搜尋。為達此目的，本研究中定義知識空間及其組成元素，並發展一組程式以產生該知識空間及知識空間搜尋機制，以提升使用者生產力。所使用到的技術包含：(1)建立知識本體，(2)計算兩詞彙同時出現頻率，(3)計算詞彙與文件關聯度，(4)發展知識空間搜尋環境。 / This study aims to provide an intelligent searching environment which users can search quickly and precisely from a large number of documents in semiconductor industry. In order to achieve the purpose, this paper defines a knowledge space and its composition elements to describe the knowledge of real world, and then develops a program to shorten the searching cost by providing the searching mechanism based on knowledge space. The techniques used in this study includes：(1) Construct 「Semiconductor Industry Ontology」(2) Compute the frequency of two terms appearing simultaneously (3) Compute the interrelatedness between terms and documents (4) Develop searching environment based on knowledge space.

文字辨識為常見的電腦視覺應用之一，隨著正確率逐漸的上升，許多新的服務相繼出現，本論文改善了筆記管理軟體最主要的問題－文字切割，並提出兩種新的中文印刷體及手寫體的分類方法。我們將筆記文件中較常見的重點標記過濾後，再使用新核心的文字結構濾波取得筆記文件中的文字區塊，新的核心數據大幅降低原始核心的計算時間。本論文也使用文字結構濾波作為分辨印刷體、手寫體的特徵值，由於文字結構濾波會依據筆畫結構給予能量回饋，使得較工整的印刷體與手寫體能有所區別，此外也使用Sobel搭配不同角度範圍進行字體辨識，實驗結果證實了本論文所提出的文字切割及字體分類方法對於筆記文件資訊的處理是有效的。 / Character recognition is an important and practical application of computer vision. With the advance of this technology, more and more services embedding text recognition functionality have become available. However, segmentation is still the central issue in many situations. In this thesis, we tackle the character segmentation problem in note taking and management applications. We propose novel methods for the discrimination of handwritten and machine-printed Chinese characters. First, we perform noise removal using heuristics and apply a stroke filter with modified kernels to efficiently compute the bounding box for the text area. The responses of the stroke filter also serve as clues for differentiating machine-printed and handwritten texts. They are further enhanced using a SVM-based classifier that employs aggregated directional responses of edge detectors as input. Experiment results have validated the efficacy of the proposed approaches in terms of text localization and style recognition.

No description available.

國民所得上升而提升了顧客購買力，物質需求不再像以往，現在同時必須滿足消費者休閒娛樂，因此結合購物以及餐飲與娛樂的大型購物中心逐漸拓展。本研究以資料採礦技術對T購物中心所提供之問卷進行資料分析，其中問卷包含了消費者基本資料、消費者行為與偏好、滿意度與建議四大部分，以統計方法分析會員與非會員之間的消費行為差異，進而做出市場區隔與行銷決策，增進顧客人數且提高消費意願和忠誠度，以及吸引非會員前來消費並申辦會員，提升顧客對T購物中心的依賴，並期望經由研究結果提供日後T購物中心於行銷計劃上之參考。 / The rising of national income promoted customers purchasing power. Material needs no longer as before,and must satisfy consumer entertainment at the same time now. Thus ,the shopping malls which combining shopping、dining and entertainment gradually expand. In this study,we used the technology of data mining to the questionnaires provided by T shopping Mall and conducted data analysis. The four parts of questionnaires contain basic information of consumers、consumer behavior、satisfaction and preferences. We analyze consumer behavioral differences between members and non-members by statistical methods and then make market segmentation and marketing decision, increase the number of customers and enhance consumer willingness and loyalty. Moreover, attract non-members to come and consume and bid for membership. Promote the dependence on T shopping mall of customers and expect the results to provide a reference on the marketing plan of T shopping mall in the future.

以往人文歷史領域的學者們，以土法煉鋼的人工方式進行資料的研究與分析，這樣的方法在資料量不大的時候還可行，但隨著數位典藏的進行以及巨量資料的興起，傳統的書本、古籍和文獻大量的數位化，若繼續使用傳統逐條分析的方式便會花費很多的時間與人力，但也因為資料數位化的關係，資訊領域的人便能利用資訊技術從旁進行協助。 而其中在觀念史研究領域裡，關鍵詞叢的研究是其中的重點之一，因為觀念可以用關鍵詞或含關鍵詞的句子來表達，所以研究關鍵詞就能幫助人文學者，了解史料文獻背後的意義與掌握當時的脈絡。因此本篇論文研究之目的在於針對收錄多篇文章的文集，探討詞彙與詞彙之間出現在文章中的情形，並利用五種的共現關係，將社群網絡的概念引入到文本分析之中，將每個詞彙當作節點，詞彙之間的關聯性當作邊建立詞彙網絡，從中找出詞彙所形成的觀念，最後實作一個由文集中探勘觀念的系統，此系統主要提供三種分析功能，分別是多詞彙觀念查詢、單詞彙觀念查詢與潛在觀念探勘。 本研究主要以《新青年》雜誌作為主要的觀察文集與實驗案例分析，《新青年》中觀念由自由主義轉向馬克思列寧主義，而我們利用本系統的確能夠找出變化的軌跡，以及探勘兩個觀念下的關鍵詞彙。 / With development of the digital archives, essays have been digitized. While it takes much time to analyze the contents of essays by human, it is beneficial to analyze by computer. This thesis aims to investigate the approach to discover concepts of essays based on social network mining techniques. While a concept can be represented as a set of keywords, the proposed approach measure the co-occurrence relationships between two keywords and represent the relationships among keywords by networks of keywords. Social network mining techniques are employed to discover the concepts of essays. We also develop the concept discovery system which provides discovery by multiple keywords, discovery by single keyword, and latent concept mining. The New Youth is taken as an example to demonstrate the capability of the developed system.

許多學者曾探討語言與音樂的互動關係。本研究則是將此議題延伸討論至漢語聖經經文入樂。本文漢語聖詩的歌詞皆截取自聖經經文，旋律的部份則是由牟維華(2007)所編。本篇主要包含了語料庫以及優選理論分析。語料庫的設置是用以觀察音節與音符的關連性、語言與音樂的邊界，以及輕聲的節奏。而優選理論分析則是藉著制約排序選出優選輸出值。本文聖詩可以兩種方式避免音符沒有對應到歌詞的現象，即一個音節對入多個音符，以及插入重複的歌詞。此兩種不同變異體可透過並存音韻(cophonology)的次語法，也就是Dep-σ 及Univormity-SM兩個制約的重新排序來解釋。由 Align-R(Long,IP)、Align-L (T, XP) 及 Align-R (T, IP) 三種制約排序，可預測出三連音邊界與歌詞的對應關係。Rhythm-N和Align-R (Long, IP) 兩個制約的排序，則可預測輕聲字的節奏長短。本研究發現音樂可被調整，以容納音節節奏。語言亦會被音樂影響，如此以符合樂譜的模組(template)。 / Scholars have discussed the interaction between language and music. This study investigates this issue in Chinese biblical hymns. The lyrics of Chinese biblical hymn discussed in this paper are from biblical verses while the melodies are composed by Mou (2007). This paper includes corpus-based and Optimality Theory analysis. Corpus is constructed to observe the phenomena of syllable-note association, language and music boundaries, and the rhythm of neutral tone syllables. In terms of Optimality Theory, the optimal output of the hymn is governed by a set of constraints. Both linking one syllable to multiple notes and inserting repeated lyrics to the hymns can avoid notes from not linking to any lyric. These two variations can be accounted for by re-ranking Dep-σ and Univormity-SM in Cophonology Theory. The constraint ranking of Align-R (Long, IP), Align-L (T, XP) and Align-R (T, IP) can predict the interaction between the tiercet boundary and lyrics. The ranking of Rhythm-N and Align-R (Long, IP) determines the rhythm of neutral tone syllables. To conclude, this study has found that music can be modified to accommodate syllable rhythm. Language can also be influenced by music to satisfy the music template.