臺灣博碩士論文加值系統

隨著多媒體應用的普及,音樂資料也大量的出現在網際網路上.而音樂的自動分類便可以把資料作分類以方便音樂搜尋的系統,也可以在音樂的分類上,可以藉由分類使得音樂搜尋系統更有效的找到資料,並且也可以找到在分類中重要的片段,提供作曲家在作曲時不同風格片段的參考.在我們的方法中,我們利用音樂內容進行音樂分類.首先我們利用網際網路收集並分析音樂分類的歌曲.我們利用節奏與旋律當作我們的分類特徵.在訓練的部分,我們呈現了將音樂資料擷取出特徵,產生重複性式樣以及評估這些式樣對於分類的重要性,也就是從這些特徵中找出對分類有用的重複性式樣,經由歌曲的分類,把重複性式樣分到各分類,並計算各式樣對分類的影響力.當要對一首歌分類時,利用人類的知覺計算式樣的相似度,最後利用計算歌曲與分類的機率決定音樂的分類.也就是找出分類特徵與重複性式樣.然後藉由這些式樣與各分類中重複性式樣的關係計算出要分類歌曲的重複性式樣與各分類的關係.進而計算出要分類歌曲與各分類的關係而決定分類的結果.在實驗的評估上,我們做了一些實驗,結果顯示根據重複性式樣的音樂分類的效果比根據音樂字串前後關係的音樂分類好,並且某些分類在某些特徵上卻有不錯的表現.例如鄉村音樂,爵士樂,藍調在旋律上,拉丁樂與搖滾樂在節奏上的正確率都有不錯的結果.

1.Introduction
2.Approach
2-1 Feature extraction and representation
2-2 Generation of significant repeating patterns
2-3 Usefulness of SRP for classification
2-4 Similarity measures for pattern matching
2-5 Class determination
3.Experiment results
3-1 Impacts of constraints
3-2 Impacts of Features
3-3 Comparison with the HMM-based approach
3-4 Impacts of evidence threshold
4.Conclusion
Reference
Appendix 1
Appendix 2

[1] C. Anagnostopoulou and G. Westermann, “Classification in Music: A Computational Model for Paradigmatic Analysis,” Proc. of the International Computer Music Conference, 1997
[2] M. Balabanovi, Y. Shoham, “Fab: Content-based Collaborative Filtering Recommendation,” communications of the ACM, March 1997, 66-72
[3] E.Battle and P. Cano, “Automatic Segmentation for Music Classification using Competitive Hidden Markov Models,” Music Information Retrieval, 2000
[4] M. Blosseville, G. Hebrail, M. Monteil and N. Penot, “Automatic Document Classification: Nature Language Processing, Statistical Analysis, and Expert System Techniques Used Together,” Proceedings of SIGIR, 1992
[5] W. Chai, and B. Vercoe, “Folk Music Classification Using Hidden Markov Models,” Proc. of International Conference on Artificial Intelligence, 2001
[6] C.C Chen and Arbee L.P. Chen, “Query by Rhythm: An approach for song retrieval in music database,” Proc. of IEEE RIDE, 1998, 139-146
[7] H.C Chen, and Arbee L.P. Chen, “A music recommendation system based on music data grouping and user interests,” Proc. of CIKM, 2001, 231-238
[8] S. Downie and M. Nelson, “Evaluation of a Simple and Effective Music Information Retrieval Method,” Proc. of SIGIR, 2000,
[9] A. Ghias, H. Logan, D. Chamberlin, and B.C. Smith, “Query by humming : Music information retrieval in an audio database,” Proc. of ACM Multimedia, 231-236
[10] S. Hagen, S. Tanja, and W. Martin, “Recognition of Music Types,” Proc. of IEEE ICASSP, 1998
[11] W.L. Hsu, S.D Lnag, “Classification Algorithms for NETNEWS Articles,” proceedings of CIKM, 1999, 114-121
[12] J.L. Hsu, C.C Liu, and Arbee L.P. Chen, “Discovering Nontrivial Repeating Patterns in Music Data,” Proc. of IEEE Transactions on Multimedia, 2001, 311-325
[13] Hisao Mase, ”Experiments on Automatic Web Page Categorization for IR system,” technical report, Stanford University, 1998.
[14] http://nzdl2.cs.waikato.ac.nz
[15] R.Papka, “Document Classification using Multiword Features,” Proc. of CIKM, 1998, 124-131
[16] D. Pogue and S. Speck, “Classical Music for Dummies,” IDG books worldwide Inc., 1999
[17] D. Roger, T. Belinda and W. David, “A machine learning approach to musical style recognition,” Proc. of International Computer Music Conference, 1997.
[18] Sniedovich, Moshe, “Dynamic programming,” Marcel Dekker, Inc., 1992
[19] Alexandra L. Uitdenbogerd and Justin Zobel, “Manipulation of Music For Melody Matching,” Proc. Of the sixth ACM international conference on Multimedia, 1998, 235-240
[20] E. Wold, T. Blum, D. Keislar and J. Wheaton, “Content-Based Classification, Search, and Retrieval of Audio,” IEEE Multimedia, 1996, 27-36