In this research, we have proposed a top-down approach for a music compositional system. Data mining techniques have been utilized to analyze and discover the common patterns or characteristics of music structure, melody style and motifs from the given musical pieces. The patterns discovered and the characteristics which constitute music structure, the melody style, and the motif selection model. The proposed system generates music based on these three models. The experimental results show that it is not easy to distinguish the system-generated music from the human-composed music. Future work includes of embedding other compositional elements such as rhythmic development, mode, and tone color into the composition process.
References
[1] R. Agrawal and R. Srikant, Fast Algorithms for Mining Association Rules, In Proc. of International Conference on Very Large Data Bases VLDB'94, 1994.
[2] R. Agrawal and R. Srikant, Mining Sequential Patterns, In Proc. of IEEE International Conference on Data Engineering ICDE’95, 1995.
[3] C. Ame and M. Domino, Cybernetic Composer: An Overview. In M. Balaban, K. Ebcioglu, and O. Laske, ed., Understanding Music with AI. AAAI Press, 1992.
[4] B. Bel, Migrating Musical Concepts: An Overview of the Bol Processor, Computer Music Journal, Vol.
22, No. 2, 1998.
[5] J. A. Biles, GenJam: A Genetic Algorithm for Generating Jazz Solos, In Proc. of International Computer Music Conference, ICMC'94, 1994.
[6] T. M. Blackwell and P. Bentley, Improvised Music with Swarms, In Proc. of the Congress on Evolutionary Computation CEC’02, 2002.
[7] A. R. Brown, Opportunities for Evolutionary Music Composition, In Proc. of Australasian Computer Music Conference ACMA'02, 2002.
[8] E. Cambouropoulos, The Local Boundary Detection Model (LBDM) and its Application in the Study of Expressive Timing, In Proc. of International Computer Music Conference ICMC'01, 2001.
[9] W. Chai and B. Vercoe, Folk Music Classification Using Hidden Markov Models, In Proc. of the International Conference on Artificial Intelligence ICAI’01, 2001.
[10] C. C. J. Chen and R. Miikkulainen, Creating Melodies with Evolving Recurrent Neural Network, In Proc.
of International Joint Conference on Neural Networks IJCNN'01, 2001.
[11] H. C. Chen, C. H. Lin, and A. L. P. Chen, Music Segmentation by Rhythmic Features and Melodic Shapes, In Proc. of IEEE International Conference on Multimedia and Expo ICME’04, 2004.
[12] D. Conklin, Music Generation from Statistical Models, In Proc. of the Symposium on Artificial Intelligence and Creativity in the Arts and Sciences AISB'03, 2003.
[13] D. Cope, The Algorithmic Composer, A-R Editions, 2000.
[14] D. Cope, Recombinant Music Using the Computer to Explore Musical Style, IEEE Computer, Vol. 24, No. 7, 1991.
[15] D. Cope, Experiments in Musical Intelligence, A-R Editions, 1996.
[16] D. Cope, Computer Modeling of Musical Intelligence in EMI, Computer Music Journal, Vol. 16, No. 2, 1992.
[17] D. C. Correa, A. L. M. Levada, J. H. Saito and J. F. Mari, Neural Network Based Systems for Computer-Aided Musical Composition: Supervised x Unsupervised Learning, In Proc. of ACM Symposium Applied Computing SAC'08, 2008.
[18] R. Dannenberg, B. Thom, and D. Watson, A Machine Learning Approach to Musical Style Recognition, In Proc. of International Computer Music Conference ICMC’97, 1997.
[19] C. T. Davie, Musical Structure and Design, Dover Publications, 1966.
[20] S. Dubnov, G. Assayag, O. Lartillot and G. Gejerano, Using Machine-Learning Methods for Musical Style Modeling, IEEE Computer, Vol. 36, No. 10, 2003.
[21] D. Espi, Ponce de Leon,P. J., C. Perez-Sancho, et al, A Cooperative Approach to Style-Oriented Music Composition, In Proc. of International Workshop on Artificial Intelligence and Music MUSIC-AI'07, 2007.
[22] M. Farbood, Analysis and Synthesis of Palestrina-Style Counterpoint Using Markov Chains, In Proc. of International Computer Music Conference ICMC'01, 2001.
[23] C. Guéret, N. Monmarché and M. Slimane, Ants Can Play Music, In Proc. of Ant Colony Optimization and Swarm Intelligence ANTS'04, 2004.
[24] M. C. Ho and M. K. Shan, Theme-based Music Structural Analysis, submitted to IEEE Multimedia, 2007.
[25] J. L. Hsu, C. C. Liu and A. L. P. Chen, Efficient Repeating Pattern Finding in Music Database, IEEE Transaction on Multimedia, Vol. 3, No. 3, 2001.
[26] N. C. Jones and P. A. Pevzner, An Introduction to Bioinformatics Algorithms, The MIT Press, 2004.
[27] I. Karydis, A. Nanopoulos and Y. Manolopoulos, Finding Maximum-length Repeating Patterns in Music Databases, Multimedia Tools and Applications, Vol. 32, No. 1, 2007.
[28] P. van Kranenburg and E. Backer, Music Style Recognition- A Quantitative Approach, In Proc. of Conference on Interdisciplinary Musicology CIM’04, 2004.
[29] O. Lartillot, S. Subnov, G. Assayag and G. Bejerano, Automatic Modeling of Musical Style, In Proc. of International Conference in Computer Music ICMC'02, 2002.
[30] Y. L. Lo, W. L. Lee and L. H. Chang, True Suffix Tree Approach for Discovering Non-trivial Repeating Patterns in a Music Object, Multimedia Tools and Applications, Vol. 37, No. 2, 2008.
[31] K. McAlpine, E. R. Miranda and S. Hoggar, Composing Music with Algorithms: A Case Study System, Computer Music Journal, Vol. 23, No.2, 1999.
[32] M. Melucci and N. Orio, A Comparison of Manual and Automatic Melody Segmentation, In Proc. of International Symposium on Music Information Retrieval ISMIR’02, 2002.
[33] E. R. Miranda, Composing Music with Computers, Focal Press, 2001.
[34] T. Miura and K. Tominaga, An Approach to Algorithmic Music Composition with an Artificial Chemistry, In Proc. of the German Workshop on Artificial Life, 2006.
[35] K. Muscutt, Composing with Algorithms: An Interview with David Cope, Computer Music Journal, Vol.
31, No. 3, 2007.
[36] G. Papadopoulos and G. Wiggins, A Genetic Algorithm for Generation of Jazz Melodies, In Proc. of Software Technology and Engineering Practice STEP'98, 1998.
[37] M. Pearce and G. Wiggins, Towards A Framework for the Evaluation of Machine Compositions, In Proc.
of Symposium on Artificial Intelligence and Creativity in the Arts and Sciences AISB'01, 2001.
[38] P. Reagan, Computer Music Generation via Decision Tree Learning, Master Thesis, Department of Computer Science, Carnegie Mellon University, USA, 1999.
[39] P. J. Ponce de Leon and J. M. Inesta, Pattern Recognition Approach for Music Style Identification Using Shallow Statistical Descriptors, IEEE Transactions on Systems, Man, and Cybernectics,-Part C:
Applications and Reviews, Vol. 37, No. 2, 2007.
[40] M. K. Shan, F. F. Kuo, and M. F. Chen, Music Style Mining and Classification by Melody, In Proc. of IEEE International Conference on Multimedia and Expo ICME’02, 2002.
[41] M. K. Shan and F. F. Kuo, Music Style Mining and Classification by Melody, IEICE Transactions on Information and System, Vol. E86-D, No. 4, 2003.
[42] A. Sorensen and A. R. Brown, Introducing jMusic, In Proc. of the Australasian Computer Music Conference, 2000.
[43] L. Stein, Structure & Style: The Study and Analysis of Musical Forms, Summy-Birchard Music, 1979.
[44] A. Takasu, T. Yanase, T. Kanazawa, and J. Adachi, Music Structure Analysis and Its Application to Theme Phrase Extraction, In Proc. of European Conference on Research and Advanced Technology for Digital Libraries ECDL’99, 1999.
[45] K. Tominaga and M. Setomoto, An Artificial-Chemistry Approach to Generating Polyphonic Musical Phrases, Lecture Notes in Computer Science, Vol. 4976, 2008.
[46] A. L. Uitdenbogerd and J. Zobel, Melodic Matching Techniques for Large Music Databases, In Proc. of ACM International Conference on Multimedia MM'99, 1999.
[47] P. E. Utgoff and P. B. Kirlin, Detecting Motives and Recurring Patterns in Polyphonic Music, In Proc. of International Computer Music Conference ICMC’06, 2006.
[48] K. Verbeurgt, M. Dinolfo and M. Fayer, Extracting Patterns in Music for Composition via Markov Chains, In Proc. of International Conference on Innovations in Applied Artificial Intelligence, 2004.
[49] T. Weyde, J. Wissmann, and K. Neubarth, An Experiment on the Role of Pitch Intervals in Melodic Segmentation, In Proc. of International Symposium on Music Information Retrieval ISMIR’07, 2007.
[50] G. Wiggins, G. Papadopoulos, S. Phon-Amnuaisuk and A. Tuson, Evolutionary Methods for Musical Composition, In Proc. of Anticipation Music and Cognition CASYS'98, 1998.
[51] I. H. Witten and E. Frank, Data Mining: Practical Machine Learning Tools and Techniques, CA: Morgan Kaufmann, 2005.
國科會補助專題研究計畫項下出席國際學術會議心得報告
日期： 年 月 日
一、參加會議經過
ACM International Conference on Information and Knowledge Management, CIKM 是 ACM 結合資料庫 (Database)、資料探勘(Data Mining)、資訊擷取(Information Retrieval)、知識管理(Knowledge Management) 領域的權威會議。歷年的 CIKM 都有不少精彩的研究成果發表。本屆會議共收到 918 篇來自全球五大 洲的論文投稿，所發表的論文包括了 134 篇 Full Paper 的 Oral Presentation、93 篇 Short Paper 的 Oral Presentation、98 篇 Poster 海報論文。其中，Full Oral Presentation Paper 的 Accepting Rate 為 15%，Short Oral Presentation Paper 的 Accepting Rate 為 10%，Poster 海報論文的 Accepting Rate 為 10%，累計 Full Paper, Short Paper, Poster 的 Accepting rate 為 35%。今年的會議為第二十屆在英國的哥拉斯哥舉行，由 格拉斯哥大學主辦。台灣今年有台大電機系陳銘憲教授、台大資工系的吳家麟教授、台大資工系的陳 信希教授、台大資工系的鄭卜任教授、連同敝實驗室共 6 篇論文在 CIKM 發表。敝實驗室此次投稿了 2 篇論文，一篇結合 Social Media 與 Music Playlist 的 Social Music Recommendation，一篇則是 Context-based People Search in Labeled Social Networks。前者很可惜在實驗部分由於資料來源的關係，
計畫編號 NSC－98－2221－E－004－007－MY2 計畫名稱 以音樂動機探勘為導向的電腦音樂作曲 出國人員
姓名 沈錳坤 服務機構
及職稱 政治大學資訊科學系教授 會議時間 100 年 10 月 24 日至
100 年 10 月 28 日 會議地點 英國格拉斯哥
會議名稱
(中文) 第 20 屆國際計算機學會資訊與知識管理會議
(英文) 20th ACM International Conference on Information and Knowledge Management, ACM CIKM, 2011.
發表論文 題目
(中文) 社群網絡上以情境為導向的人名搜尋
(英文)Context-based People Search in Labeled Social Networks 附件四
不夠完整。後者則獲評審肯定。
今年的 Keynote speech 包括 MIT 的 Professor David R. Karger 主講 Creating User Interfaces that Entice People to Manage Better Information、University of Melbourne 的 Professor Justin Zobel 主講 Data, Health, and Algorithmics: Computational Challenges for Biomedicine、來自義大利 Università di Roma La Sapienza 的 Professor Maurizio Lenzerini 主講 Ontology-based Data Management。與我的研究關係密切的是 Professor David R. Karger 的主題演講。雖然對於這些議題都不陌生，但 Prof. Karger 完整的整理與介紹 令人對於相關議題有整合而嚴謹的體會。尤其，對於 Collaborative Filtering, Social Media 有精闢的分析。
他也點出幾個重要研究議題。對於我們正研究的 Social Computing、Social Music Recommendation 之研 究有很大的幫助。
我的報告安排在 25 日下午的 Query Answering and Social Search 這 Session。Session Chair 是來自於 Indian University 的 Prof. Yuqing Wu。Prof. Wu 的研究專長在 XML Database，尤其是 XML 的 Query Processing。之前曾研讀過其相關論文，其實驗室發展的系統也有在這次會議的 Demo。這場主要的論 文都偏重在 Search。我們發表的論文主要是利用 Social Context 協助使用者在 Social Networks 上的搜 尋，我們將此 Social Search Problem 轉換為 Team Formation 的問題，並提出在大量 Social Networks 上 的演算法。我們的報告獲得在場學者踴躍的討論，大家對於以 Context 輔助 Social Network 上的搜尋很 感興趣，唯一感到可惜的是有位學者誤解我們的研究是解決 Namesake 的問題。
另一場印象深刻的主題是 26 日下午的 Session: Social Networks and Communities。尤其第一篇發表 的論文: Discovering Top-k Teams of Experts with/without a Leader in Social Networks，與我近年的研究興 趣息息相關。雖然作者沒有出席，而是由另一篇論文的作者代為報告，但其問題定義的變形非常有創 意。很可惜的是可能因為原作者沒出席，因此沒有獲得現場學者的關注。
今年在會議的前一天與後一天分別有 10 場 Tutorial 與 15 場 Workshop。雖然 Tutorial 中包括來自於 Yahoo! Research 的 Information Retrieval Challenges in Computational Advertising、UC Santa Barbara 的 Information Diffusion in Social Networks: Observing and Affecting What Society Cares About、與 Big Data 相關的 Large-Scale Array Analytics: Taming the Data Tsunami，但受限於經費限制而沒能參加。
二、與會心得
ACM CIKM 是結合 Data Base, Data Mining, Information Retrieval, Knowledge Management 跨領域的
權威會議。由每年發表的論文可以觀察這些領域的研究發展趨勢。這幾年逐漸可以觀察出產業界的大 量資料對於這些領域的影響。除了 Microsoft Research, Google Research, Yahoo! Research 因為有大量使 用者資料而有不少研究論文發表，很多研究的實驗也都以大量資料來驗證。
此外，27 日的 Industrial Event 中我印象最深刻的是來自於 Google 的 Ed Chi 介紹 Model-Driven Research in Social Computing。這與我目前有關 Social Network 有密切的關係，Ed Chi 點出目前 Social Network 上有關 Information Diffusion 的 Theory of Influential 的盲點，並提出一些 Model-Driven 的 Issues。
這次會議如同 ACM KDD, IEEE ICDM，逐漸有 Social Network 的相關論文，只是這方面的成長數量 沒有其他會議明顯。希望明年會議將會有更多結合 Social Network 的相關論文。
三、考察參觀活動(無是項活動者略)
這次會議的舉行地點雖然是在格拉斯哥的 Crowne Plaza，但基於對亞當史密斯的崇仰，特別到主辦 單位格拉斯哥大學參觀。除了看到亞當史密斯的雕像，也參觀了大學的博物館。對於格拉斯哥大學在 啟蒙時代，所扮演的角色與對社會的影響，印象深刻。同時，由倫敦回國前一天，也拜訪了 University of College London (UCL)的 Cancer Institute，因為敝實驗室畢業的系友(大學部專題學生，後來在中研院 TIGP 獲得博士學位)詹博士正在此進行有關生物資訊的博士後研究。與其詳談研究工作情形，深深體 會到 Data Science 在 Bioinformatics 扮演的角色。
四、攜回資料名稱及內容
ACM CIKM 2011 論文集、隨身碟、Springer, Oxford 新書介紹
Context-based People Search in Labeled Social Networks
Cheng-Te Li
Graduate Institute of Networking and Multimedia, National Taiwan University, Taipei, Taiwan
d98944005@csie.ntu.edu.tw
Man-Kwan Shan
Department of Computer Science, National Chengchi University, Taipei, Taiwan
mkshan@cs.nccu.edu.tw
ABSTRACT
In online social networking services, there are a range of scenarios in which users want to search a particular person given the targeted person one’s name. The challenge of such people search is namesake, which means that there are many people possess the same names in the social network. In this paper, we propose to leverage the query contexts to tackle such problems. For example, given the information of one's graduation year and city, the last names of some individuals, one may wish to find classmates from his/her high school. We formulate such problem as the context-based people search. Given a social network in which each node is associated with a set of labels and given a query set of labels consisting of a targeted name label and other context labels, our goal is to return a ranking list of persons who possess the targeted name label and connects to other context labels with minimum communication costs through an effective subgraph in the social network. We consider the interactions among query labels to propose a grouping-based method to solve the context-based people search. Our method consists of three major parts. First, we model those nodes with query labels into a group graph which is able to reduce the search space to enhance the time efficiency.
Second, we identify three different kinds of connectors which connecting different groups, and exploit connectors to find the corresponding detailed graph topology from the group graph.
Third, we propose a Connector-Steiner Tree algorithm to retrieve a resulting ranked list of individuals who possess the targeted label. Experimental results on the DBLP bibliography data show that our grouping-based method can reach the good quality of returned persons as a greedy search algorithm at a considerable outperformance on the time efficiency.
Categories and Subject Descriptors
H.2.8 [Database Management]: Database Applications – Data mining. H.3.3 [Information Storage and Retrieval]: Information Search and Retrieval – Information Filtering.
General Terms
Algorithms, Performance, Design.
Keywords
People Search, Social Network, Context-based Search.
1. INTRODUCTION
In online social networking services such as Facebook, Twitter, and LinkedIn, it is essential to provide people search which searches for an individual by name. However, if the query name is a namesake, especially if there exist millions of individuals share the query name, it would be difficult to find the target person over social networking services.
One approach is the context-based people search which search for an individual not only by name of the target, by also by the social contexts of the target. The user specified social contexts may be the first names of target’s friends, the last names of target’s classmates, the hobbies of the target’s colleagues, hometowns of target’s relatives, and so on. These social contexts are labels associated with each individual in social networking services.
The idea of the proposed context-based social search is illustrated by Figure 1. Each person is associated with labels, where some have many labels and others provide few labels or only their user
The idea of the proposed context-based social search is illustrated by Figure 1. Each person is associated with labels, where some have many labels and others provide few labels or only their user