一、研究目的 1.實務性的目的：研究一貫化與分段化採購作業方式在中信局購料處實施後所產生的作業績效差異，衡量其優劣，探討其原因，以尋求較佳之採購作業方式提高作業績效，節省外匯，增進國家建設潛力。 2.理論性目的:以實例調查的結果為驗證，探討作業方式與採購作業間之關係，進而推論作業方式與作業性質間之關係及分工作業方式之選擇適用原則。 二、研究方法 本研究以實例配合理論的方式來研究本文的主旨。先以實例中不同作業方式的利害關係人為對象，調查其對不同作業方式下的態度性反應，而據以衡量不同作業方式下作業績效之優劣。其次從另一角度以訪談的方式探尋作業組織內人員之心理反應是否與調查的結果相一致。最後再以理論依據來分析、推論其應有的結果而與調查結果相對照，以彼此證明其正確性，而可將該理論依據用於其他之推論上。 三、研究內容及結構 本研究共分為八章，第一章緒論，說明本研究之目的、意義及研究的觀念性模型。第二章描述與本研究有關的背景資料，採購一般概況及中信局購料處之組織結構、作業內容。第三章探討本研究所依據的理論模型，包括「影響作業績效之理論模型」及「分工原理適用模型」，以作為分析解釋之依據。第四章以中信局購料處為實例，調查其作業績效受不同作業方式影響的情形，以作為支持本研究目的的實證依據。第五章以從另一角度的訪談資料來以事實現象說明造成調查結果的原因。第六章從理論模型的觀點來分析採購作業實施分段作業對作業績效可能的影響，以便與調查結果相對照、相證實，進而據其作粗略性的推論。第七章說明本研究結果應用上的限制及選擇作業方式時的觀念性模型。第八章首先重述本文之研究過程，並據其研究過程中，所得到的結果，作成結論及建議。 四、結論 分段化作業方式在中信局購料處實施的結果，共作業績效較一貫化作業方式為差。分段化作業方式可能不適宜集中採購作業時使用。在其他與採購作業內容相類似的行業裡可能亦不適宜用分段化作業方式。

由於電影工業的蓬勃發展，以及數位化視訊分析與儲存技術的進步，使用者可藉由DVD所提供的故事分段索引來快速瀏覽及搜尋影片。因此一套能對電影作故事分段的工具是不可或缺的。 / 本論文的研究目的是針對人際關係類型的電影做故事單元的分段。我們提出以角色的社會網絡為基礎的方法，作電影故事單元的分段。此方法包括四個步驟。首先對電影作場景變換偵測。接著，我們利用人臉辨識技術擷取出每一個場景出現的角色。第三，我們考慮角色重要性對分段的影響，利用社會網絡分析中計算角色的網絡中心性，來衡量角色的重要性。最後，以角色為特徵值，比對場景之間的角色來計算相似度，並且利用循序性叢集分析，來達到電影故事單元的分段。我們的實驗針對四部人際關係類型的電影，以切成故事單元來評估分段的效果，實驗顯示以角色的社會網路為基礎的方法，準確率介於0.63到0.94之間。 / With the progress of entertainment industry, and the advances of digital video analysis and storage technologies, users can utilize the indexes of the DVD chapters for quick access, retrieval and browsing of movie content. Therefore, development of automatic movie content analysis is important. In this thesis, we focus on the romance and relationship movies, which contain the narrative of the relation between peoples. We propose a novel method for film narrative exploration based on social network analysis of characters. There are four steps. First, we perform movie scene change detection to segment a movie into scenes. In the second step, we extract the characters as the feature model of scenes by utilizing the face recognition system. Then, we measure the weight value of the characters by the centrality of social network analysis. Finally, we calculate the cosine similarity between scenes, and segment a movie into story units by using sequential clustering algorithm. We conduct experiments on four romance and relationship movies. Experimental result show that our proposed story unit segmentation method based on social network analysis of characters achieves from 63% to 94% accuracy.

抽取出音樂當中的近似重複樣式來做音樂檢索可以減少要比對的資料量，但是使用者若使用沒有重複的旋律來查詢便會有找不到歌曲的情況。另一方面，將音樂分段成phrase可以減少樹狀索引結構的空間，亦可減少查詢處理時間，但是使用者的查詢若是跨越phrase的，也將影響查詢結果。 在本論文中，我們比較了以近似重複樣式與phrase兩種不同的音樂特徵用來做音樂檢索的效果以及效率。根據實驗顯示，使用者的查詢是重複旋律的機會大於單一phrase，所以用近似重複樣式作為音樂查詢比對資料效果是比phrase好的。而在1-D List索引結構下，近似重複樣式的效率也優於phrase。除此之外，本論文也提出了一個新的近似重複樣式抽取方法，實驗證明我們的方法是有效的。 / Extract the approximate repeating pattern from music data will decrease the volumes of music data that need to be tested when music retrieve. If the user’s query is not a repeating melody, it can’t retrieve the music that the user wants correctly. In addition, segment the music by phrase will decrease the space that tree-like index structure need, and also decrease the retrieval processing time. If the user’s query is not a single phrase, it will influence the effectiveness of retrieval. In this thesis, we compare the effectiveness and efficiency of music retrieval methods with two different music features (approximate repeating pattern and phrase). According to experiment results, the probability that user’s query is repeating melody is more than the probability that user’s query is a single phrase. Therefore, we are of the opinion that the effectiveness that use approximate repeating pattern to process retrieval is more prominent than the effectiveness that phrase to process retrieval. Furthermore, the efficiency that use approximate repeating pattern to process retrieval is more outstanding than use phrase under 1-D List index structure. Besides, a new approximate repeating pattern extraction method is proposed. Experiment results show that our approximate repeating pattern extraction method can work correctly.

音樂分段在研究音樂分析相關的領域是很重要的研究題目。音樂的分段可以提供作音樂結構分析、音樂瀏覽、音樂內容查詢與音樂摘要等應用。本論文的研究目的就是對音樂作自動分段，以幫助使用者能快速瀏覽音樂的內容。因此，我們針對音樂的主題作主題式的分段。 音樂的主題是取決於作曲者的動機，動機是構成音樂主題的基本因素。為了能夠以主題為基礎作音樂分段，我們必須找出決定音樂主題的因素。動機會有規則性的出現在整首音樂當中，所以我們可以利用動機出現的規則來探勘音樂的動機。 我們提出一個以主題對音樂作分段的方法，總共分為四個主要的步驟。第一，我們從原始的音樂資料擷取出主旋律的部分。第二，將主旋律做粗略分段。我們利用探勘Non-trivial重複樣式的技術[17]，來找出粗略段落。第三，從粗略段落中探勘動機。我們利用Stein所提出來的動機變化規則，修改傳統探勘重複序列的方法，做動機的探勘。最後，我們利用探勘出來的動機對主旋律作精細分段。我們針對MIDI音樂檔案利用提出來的方法，實做出一個系統，找出音樂的主題段落。 先前研究在評估實驗結果時，多採用Precision與Recall去評估實驗的結果。然而，這樣的評估方法並不能表現出實驗結果與正確答案之間的相似程度。所以我們提出新的評估方法，根據實驗結果與正確答案之間的相似程度來評估實驗的準確率。根據實驗結果顯示，我們的方法準確率約65%。 / Music segmentation is one of the important issues in music analysis. Music segmentation can be utilized for music structure analysis, music browsing, content-based music retrieval, and music summarization. In this theis, we proposed a music segmentation method based on the music theme to provide users the capability to browse music segments by theme. Motives, the concepts of the composer, are the basic elements of music themes. Music themes were constructed by motives. In order to segment music by themes, we have to discover motives. Most motives repeated in the music by some motivic treatment rules. Therefore, motives can be discovered by these rules. We proposed the theme segmentation method. There are four steps. Firstly, we extract main melody from original music. In the second step, rough segments are generated from main melody by mining non-trivial repeating patterns. Then, motives are detected from rough segments. We modify the mining algorithm for discovering frequent patterns by applying motivic treatment rules proposed by Stein. Finally, we segment main melody based on the generated motives. Moreover, a system for segmentation of music in MIDI format was implemented. Concerning the effectiveness evaluation of music segmentation, precision and recall are used in previous research. We proposed an effectiveness measure and corresponding algorithm to evaluate the accuracy of music segmentation. Experimental results show that our proposed music segmentation method achieves 65% accuracy.

本研究旨在探討分段式評量教學法對高二學生於學習數學時學習成就的影響。研究對象為台北縣某國立高中二年級理組學生，分為實驗組42位及控制組44位共86位學生，以20週的時間進行實驗，觀察分析其學習成就的改變。 本研究結果發現，在相同教學時間下： 1. 分段式評量教學法在整體學習成就方面有正面影響且結果達顯著差異。 2. 對不同學習風格學習成就之正面影響雖未達顯著水準，但學習成就相對改善值似乎有增加的趨勢。 3. 對不同學習程度分組而言，中分組與低分組學生之學習成就方面有正面影響且達顯著差異。 4. 實施分段式評量教學法的學生比使用傳統教學法的學生在學習態度方面似乎較不會有放棄數學的現象。 綜而論之，分段式評量教學法可提供教學者，在面對數學學習成就低落的學生一個有效的引導方法，讓這些學生不僅不會放棄數學，還能漸漸的建立良好的學習習慣。 / This research mainly aims at evaluating how divided assessment teaching method would effect junior students’ mathematics-learning performance in high school. A case study was conducted on science-team junior students in a Taipei-county high school, composed of 42 students in experimental group and 44 ones in control group respectively, amounting to 86. This experiment spanned as long as 20 weeks for analysis on how students’ learning performance would be benefited. It is thus concluded in this research after evaluating both 2 group’s learning performance in term of equal length of time as below: 1. Divided assessment teaching method would have positive effect on learning performance at significance level. 2. Although divided assessment teaching method has positive effects on learning performance for various learning style subgroups, these positive effects are not significant. Their relative improvements of learning performance seemed to be increased, too. 3. When evaluated in term of original-performance level, students’ learning performance in average-level and in inferior-level subgroups both would be benefited positively at significance level. 4. Students taking divided assessment teaching method would have more persistent learning attitude than those taking traditional teaching method. In summary, divided assessment teaching method could help teachers to offer more effective teaching-guidance to students who had inferior learning performance. As a result, students would not only persist in mathematics learning but also cultivate enthusiastic learning attitude gradually.

近幾年來，華語流行音樂的發展越來越多元，而大眾所接收到的資訊是流行音樂當中的組成元素”曲與詞”，兩者分別具有賦予人類感知的功能，使人能夠深刻體會音樂作品當中所表答的內容與意境。然而，作曲與作詞都是屬於專業的創作藝術，作詞者通常在填詞時，會先對樂曲當中的結構進行粗略的分析，找出整首曲子的曲式，而針對可以填詞的部份，再進行更細部的分析將詞填入最適當的位置。流行音樂當中，曲與詞存在著密不可分的關係，瞭解歌曲結構不僅能降低填詞的門檻，亦能夠明白曲子的骨架與脈絡;在音樂教育與音樂檢索方面亦有幫助。 本研究的目標為，使用者輸入流行音樂歌曲，系統會自動分析出曲子的『曲式結構』。方法主要分成三個部分，分別為主旋律擷取、歌句分段與音樂曲式結構擷取。首先，我們利用Support Vector Machine以學習之方式建立模型後，擷取出符號音樂中之主旋律。第二步驟我們以”歌句”為單位，對主旋律進行分段，對於分段之結果建構出Self-Similarity Matrix矩陣。最後再利用Non-Negative Matrix Factorization針對不同特徵值矩陣進行分解並建立第二層之Self-Similarity Matrix矩陣，以歧異度之方式找出曲式邊界。 我們針對分段方式對歌曲結構之影響進行分析與觀察。實驗數據顯示，事先將歌曲以歌句單位分段之效果較未分段佳，而歌句分段之評測結果F-Score為0.82;將音樂中以不同特徵值建構之自相似度矩進行Non-Negative Matrix Factorization後，另一空間中之基底特徵更能有效地分辨出不同的歌曲結構，其F-Score為0.71。 / Music structure analysis is helpful for music information retrieval, music education and alignment between lyrics and music. This thesis investigates the techniques of music structure analysis for Chinese popular music. Our work is to analyze music form automatically by three steps, main melody finding, sentence discovery, and music form discovery. First, we extract main melody based on learning from user-labeled sample using support vector machine. Then, the boundary of music sentence is detected by two-way classification using support vector machine. To discover the music form, the sentence-based Self-Similarity Matrix is constructed for each music feature. Non-negative Matrix Factorization is employed to extract the new features and to construct the second level Self-Similarity Matrix. The checkerboard kernel correlation is utilized to find music form boundaries on the second level Self-Similarity Matrix. Experiments on eighty Chinese popular music are performed for performance evaluation of the proposed approaches. For the main melody finding, our proposed learning-based approach is better than existing methods. The proposed approaches achieve 82% F-score for sentence discovery while 71% F-score for music form discovery.

在迴歸分析中，若變數間具有非線性 (nonlinear) 的關係時，B-Spline線性迴歸是以無母數的方式建立模型。B-Spline函數為具有節點(knots)的分段多項式，選取合適節點的位置對B-Spline函數的估計有重要的影響，在希望得到B-Spline較好的估計量的同時，我們也想要只用少數的節點就達成想要的成效，於是Huang (2013) 提出了一種選擇節點的方式APLS (Adaptive penalized least squares)，在本文中，我們以此方法進行一些更一般化的設定，並在不同的設定之下，判斷是否有較好的估計效果，且已修正後的方法與基於BIC (Bayesian information criterion)的節點估計方式進行比較，在本文中我們將一般化設定的APLS法稱為GAPLS，並且經由模擬結果我們發現此兩種以B-Spline進行迴歸函數近似的方法其近似效果都很不錯，只是節點的個數略有不同，所以若是對節點選取的個數有嚴格要求要取較少的節點的話，我們建議使用基於BIC的節點估計方式，除此之外GAPLS法也是不錯的選擇。 / In regression analysis, if the relationship between the response variable and the explanatory variables is nonlinear, B-splines can be used to model the nonlinear relationship. Knot selection is crucial in B-spline regression. Huang (2013) propose a method for adaptive estimation, where knots are selected based on penalized least squares. This method is abbreviated as APLS (adaptive penalized least squares) in this thesis. In this thesis, a more general version of APLS is proposed, which is abbreviated as GAPLS (generalized APLS). Simulation studies are carried out to compare the estimation performance between GAPLS and a knot selection method based on BIC (Bayesian information criterion). The simulation results show that both methods perform well and fewer knots are selected using the BIC approach than using GAPLS.

在音樂中，組曲是一種特別的創作形式。組曲將多首音樂段落組合排列，並且在音樂段落之間加入間奏，形成一首音樂組曲。組曲的編曲重點在於音樂段落的編排順序及段落之間的連結。平時在宴會、舞會、餐廳、賣場等場合中，往往都會連續播放多首流行音樂。利用電腦編曲自動產生流行音樂組曲，將可提升播放音樂的銜接與流暢感。 因此，本研究利用資料探勘技術及音樂編曲理論，將多首音樂重新改編成一首組曲。系統首先將每首音樂分段並找出每首音樂的代表段落。接著，系統根據代表段落間的相似度編排順序。最後，為了達到組曲中音樂段落連接的流暢性，我們以模型訓練的方式在段落連結間加入間奏。系統從訓練資料學習產生旋律發展、和弦進程與節奏的模型，接著分析代表段落的動機、旋律、和弦及節奏，使得組曲編曲後的段落連結更為流暢且完整。本研究以流行音樂鋼琴伴奏曲為測試資料，我們分別邀請三十四位受過音樂訓練與未受音樂訓練的測試者，針對本研究所提出的鋼琴伴奏節奏辨識、代表段落萃取、段落順序編排及間奏產生，評估其效果。實驗結果顯示，本研究所提出的順序編排與間奏產生技術，對於組曲的流暢感，有著相當大的幫助。 / In music, a medley is an organized piece composed from segments of existing pieces. Ordering and bridge for connection between segments are the important elements for medley arrangement. Automatic medley arrangement is helpful for playing a set of music continuously which is common in banquet, party, restaurant, shopping mall, etc.. This thesis aims to develop the automatic medley arrangement method by using data mining techniques and music arrangement theory. The proposed method first segments each music and discovers the significant segment from each music. Then, the linear arrangement based on the similarities between significant segments is generated. Finally, in order to connect segments smoothly in the medley, the bridge between two segments is generated and inserted by using model training. Three models, melody progression, chord progression and rhythm models are learned from training data. For the experiments, testing data is collected from popular piano music and thirty-four people are invited to evaluate the effectiveness of the rhythm recognition of accompaniment, the extraction of significant segment, the linear arrangement of segments, and the creation of bridge. Experimental results show that the proposed medley arrangement method performs well.