In recent years, the music has become popular due to the evolution of the technology. Various kinds of music around us become complexities and huge. The explosive growth in the music has generated the urgent need for new techniques and tools that can intelligently and automatically transform the music into useful information. Many researches consider the music object as an continuously discrete note in time order. Repeating patterns are some subsequences which appear frequently in the music sequence. The repeating patterns usually can represent the theme of a music object. Moreover, it also can be utilized in music classification. Many methods have been proposed for finding the repeating patterns in music objects, for example, the M2P (Mining Maximum-length Patterns) method. It constructs a directed graph and uses the depth-first search to traverse the graph. It calculates the paths by the string matching algorithm to decide whether they are repeating pattern, and finds out the maximum-length repeating pattern in a music sequence. Although the M2P method is a straightforward method to find out the patterns, it consumes time in creating too many candidate patterns and performing the string matching algorithm. Therefore, in this thesis, we propose the PJ (Position-Join) method to efficiently find out the maximum-length repeating pattern. In the constructing graph step, we find out that we can modify the information in the graph, and avoid to use the string matching algorithm to decide whether a path is repeating pattern. We record the positions of length two repeating patterns in the matrix. While traversing the graph, we calculate the frequency by the information of positions. Moreover, we record the repeated path by the positions. We create terminal edges, and record the information of paths which have been traversed. We dynamically modify the graph by terminal edges. It can avoid to traverse some paths repeatedly in traversing the graph step. From our performance study based on the synthetic data and real music data, we show that our proposed PJ method is more efficient than the M2P method.

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.

Efterfrågan på matematiskt kunnande ökar i samhället. Algebraiskt kunnande gynnar lärandet inom andra delar av matematiken. Algebra innebär att upptäcka matematiska strukturer och kunna generalisera dessa. Mönsterundervisning ses som en säker väg till algebra. Samtidigt visar forskning kopplat till upprepade mönster att de flesta elever inte kan generalisera upprepade mönster innan de börjar skolan, och menar att det finns ett behov av en strukturerad undervisning kopplad till upprepade mönster. Stora delar av matematikundervisningen i Sverige utgår från matematikböcker. Därför syftar denna studie till att beskriva de upprepade mönster elever kan möta genom ett läromedel och därigenom dra slutsatser om vad elever får för möjlighet att lära om algebra genom dessa. Följande frågor besvaras: Vilken variation och omfattning av mönster och mönsteruppgifter får eleverna möta i matematikböckerna kopplade till upprepade mönster? och Vilken ytterligare vägledning ger lärarhandledningen för att elever ska utveckla förståelse av upprepade mönster?  Den teoretiska ansats studien utgår från är tidigare forskning om undervisning i växande mönster. Studien är avgränsad till matematikböcker för år 1–3. Urvalet av material innefattar 6 matematikböcker i serien Koll på matematik samt tillhörande lärarhandledningsböcker. Samtliga böcker har analyserats för att finna upprepade mönster. Dessa mönster sorterades efter kategorierna riktning, uttrycksformer, mönsterstruktur och uppgifts typ. Därefter genomsöktes lärarhandledningen efter förslag på undervisning kopplade till de upprepade mönstren som fanns i elevböckerna. Resultatet visar att det främst finns en variation på uttrycksformer och mönsterstrukturer i matematikböckerna och en mindre variation på riktning. Uppgiftstyperna varierar i viss mån. Vidare visar resultatet att lärarhandledningen tillförde flera kommunikativa aspekter kopplade till mönster och även fler metoder vilket inte framgick i elevböckerna. / The request for mathematical knowledge is increasing in today's society. Algebraic knowledge favour further learning in other areas of mathematics. Algebra involves finding mathematical structures and being able to generalize these. Teaching about patterns are described as a safe path to learn algebra. At the same time, research on repeating patterns proves that most children cannot make generalization about repeating patterns before they enter school, therefore structured teaching connected to repeated patterns is needed. Teaching in Sweden is mostly based on mathematics schoolbooks. Therefore, the aim for this study is to describe the repeated patterns students can encounter through teaching material and thereby draw conclusions about which opportunities students have to learn about algebra through these. The questions answered in this study are: Which variation and amount of patterns and pattern tasks do the pupils get exposed to in the mathematics books connected to repeating patterns? and What does the additional teacher guides provide with for pupils understanding of repeating patterns? The study is limited to mathematics schoolbooks for schoolyear 1-3.  The theoretical approach in this study is based on previous research on teaching methods connected to repeated patterns. The selection of mathematics books includes 6 mathematics books from the teaching materials Koll på matematik. All pages have been scanned to find repeated patterns. The repeating patterns were categorized by the pattern's direction, different expression of pattern, structure, and task type. Afterwards, the teacher's manual was searched for teaching suggestions linked to the repeating patterns that were found in the student books. The result indicates that the variation is primally connected to different expression of pattern and pattern’s structure in the mathematics books and that there are minimal variation of the patterns direction. The task types vary to a certain extent. Furthermore, the result indicate that the teacher's guide contributes with several communicative aspects connected to repeating patterns and more methods that were not apparent in the student books.

現在有許多對於眼動軌跡與人在認知方面的研究，包括理解狀態以及感興趣的程度；其中，閱讀文章時的眼動軌跡是最常被討論及研究的題材。而本研究的目的就是希望探討讀者在閱讀時的眼動軌跡，與其感興趣程度之間是否存在關係。 / 本研究的特色在於，我們不用一般分析眼動時關心每個AOI（area of interest）上的眼動資料，而是希望將眼動資料以序列的方式分析，並且運用資料探勘的方法，找出眼動序列中區分感興趣程度的眼動軌跡特徵的片段。 / 透過對於眼動軌跡的分析，我們希望研究的結果，在未來可以運用在資訊檢索的領域上，成為一種有效的「隱含式回饋（implicit feedback）」的方式，以改善現有資訊檢索效能。 / Much research has been performed on the relationship between eye movements and human cognition, including comprehension and interesting degree. The purpose of our research is to find out if there are relationships between eye movements of reading and interesting degree. / Instead of analyzing the eye movements on each area of interest, the characteristic of our research is to transform eye movements to sequence data, and to determine the eye movement patterns which discriminate whether user is interesting or not by using the method of data mining. / Through the analysis of the eye movements, our research result can be used as one way of implicit feedback of information retrieval to improve the effectiveness of the search engine.

華語流行歌詞一直是語言、文學、音樂或是文化研究等相關科系赤手可熱的研究題目，內容包含作詞者、修辭分析、風格、用韻及語言表達等，然由於歌詞數量龐大，難以全部以人工分析。 近年來，資訊科技日新月異不斷地進步，隨著Big Data議題受到注目，Data Mining在近年來相當熱門，然而針對華語流行歌詞的巨量資料探勘與分析研究並不多。因此，本論文研究以程式來自動化分析歌詞的樣式與特性，包括詞彙頻率、詞彙相鄰關係分析、歌名分析、使用語系分析、舊曲新唱、歌詞風格自動分類、用韻及修辭等，而研究資料係透過網路擷取知名網站內容，包含魔鏡歌詞網 (Mojim.com)、臺北之音HitFM聯播網 (www.hitoradio.com) 及教育部重編國語辭典，透過分析規則及以Non-Trivial Repeating Pattern等方法，來完成分析及系統實作。 透過華語流行歌詞的大量分析，探勘及了解各種歌詞的風格與特性，將可了解各種歌詞、作詞者的風格與特色，進而應用在歌詞資料的管理與查詢。 此外，本研究將八萬多首歌詞的各種分析資料設置成網站，提供予學術研究使用，希冀此研究資料能使華語流行歌詞相關研究研究，進行更深入地探討。 / Chinese popular music lyrics has been a popular topic for researchers who major in languages and literature, music or culture. Related studies include of lyricists, rhetoric methods, styles, rhyme and language expression. However, all these studies were performed by manual analysis. It is difficult to analyze large amount of lyrics manually. With advances in computer technology, big data and data mining techniques have been widely used in different kinds of data. However, to the best of our knowledge, none have been done on pattern mining from big data of lyrics of Chinese popular music. Therefore, the objective of this thesis is to discover patterns from tremendous lyrics data based on data mining techniques. We use data downloaded from www.mojim.com, http://dict.revised.moe.edu.tw/cbdic/ and http://www.hitoradio.com (Hit FM). Data mining methods are employed to find lyrics’ patterns and features, including frequent words, word adjacency, analysis of hit songs' names, lyrics’ language studies, cover song research, automatic style prediction, rhyme and rhetoric patterns. With the analysis of tremendous lyrics and data, the developed approaches of this thesis will be helpful for discovering distinguishing styles of lyrics and lyricists.

目前大部分的聽眾主要是透過歌詞與樂曲的搭配來了解音樂所要表達的內容，因此歌詞創作在目前的音樂工業是很重要的一環。一般流行音樂創作是由作曲人與作詞人共同完成，然而有另一種方式是將既有的詩詞做為歌詞，接著重新譜曲的方式產生新的流行音樂。這種創作方式是讓舊有的詞或曲注入新的生命力，得以流傳到現在。因此本研究希望可以為一首旋律推薦適合配唱的歌詞，以對數位音樂達到舊曲新詞的加值應用。本論文包括兩個部分，分別為：(1)自動分析歌詞的詞式，找出每個段落的位置與其段落的標籤；(2)詞曲結構搭配，找出相符合結構的詞與曲，並且同步每個漢字與音符。 本論文的第一部分為詞式分析，首先將歌詞擷取四個面向的特徵值，分別為(1)句字數結構；(2)拼音結構；(3)詞性；(4)聲調音高。第二步驟，利用這四種特徵值分別建立詞行的自相似度矩陣(Self Similarity Matrix)，並且利用這四個特徵的自相似度矩陣產生一個線性組合自相似度矩陣。第三步驟，建立在自相似度矩陣上我們做段落分群以及家族(Family)組合找出最佳的分段方式，最後將找出的分段方式利用我們整理出來的規則讓電腦自動標記段落標籤。第二部分為詞曲結構搭配，首先我們將主旋律的樂句以及歌詞的詞句做第一層粗略的對應，第二步驟，將對應好的樂句與詞句做第二層漢字與音符細部的對應，最後整合兩層對應的成本當做詞曲搭配的分數。 我們以KKBOX音樂網站當做歌詞來源，並且請專家標記華語流行歌詞資料庫的詞式。實驗顯示詞式分析的Pairwise f-score準確率達到0.83，標籤回復準確率達到0.78。詞曲結構搭配中，查詢的歌曲其原本搭配的歌詞，推薦排名皆為第一名。 / Nowadays, lots of pop music audiences understand the content of music via lyrics and melody collocation. In general, a Chinese pop music is produced by composer and lyricist cooperatively. However, another producing manner is composing new melody with ancient poetry. Therefore, we want to recommend present lyrics for a melody and then achieving value-added application for digital music. This thesis includes two subjects. The first subject is lyrics form analysis. This subject is finding the block of verse, chorus, etc., in lyrics. The second subject is structure alignment between lyrics and melody. We utilize the result of lyrics form analysis and then employ a 2-tier alignment to recommend present lyrics which is suitable for singing. In lyrics form analysis, the first step, we investigate four types of feature from lyrics: (1) Word Count Structure; (2) Pinyin Structure; (3) Part of Speech Structure; (4) Word Tone Pitch. For the second step, we utilize these four types of feature to construct a SSM(Self Similarity Matrix), and blend these four types of SSM to produce a linear combination SSM. The third step is clustering blocks and finding the best Family combination based on SSM. Finally, a rule-based technique is employed to label blocks of lyrics. For the second subject, the first step is aligning music phrases and lyrics sentences roughly. The second step is aligning a word and a note for corresponding phrase and sentence. Finally, we integrated the cost of two-level alignment regarded as the lyrics and melody collocation score. We collect lyrics from KKBOX, a music web site, and invite experts label ground truth of lyrics form. The experimental result of lyrics form analysis shows that the proposed method achieves the Pairwise f-score of 0.83, and the Label Recovering Ratio of 0.78. The experiment of structure alignment between lyrics and melody shows that the original lyrics of query melodies are ranked number one.