請用此 Handle URI 來引用此文件：
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/21183完整後設資料紀錄
| DC 欄位 | 值 | 語言 |
|---|---|---|
| dc.contributor.advisor | 陳宏銘 | |
| dc.contributor.author | Yi-Hsun Lin | en |
| dc.contributor.author | 林奕勳 | zh_TW |
| dc.date.accessioned | 2021-06-08T03:28:17Z | - |
| dc.date.copyright | 2019-12-25 | |
| dc.date.issued | 2019 | |
| dc.date.submitted | 2018-08-17 | |
| dc.identifier.citation | [1] O. Kanishcheva and G. Angelova, “A pipeline approach to image auto-tagging refinement,” in Proc. BCI, Craiova, Romania, 2015: ACM, p. 9.
[2] T. Bertin-Mahieux, D. Eck, F. Maillet, and P. Lamere, “Autotagger: A model for predicting social tags from acoustic features on large music databases,” Journal of New Music Research, vol. 37, no. 2, pp. 115–135, 2008. [3] E. Law, B. Settles, and T. Mitchell, “Learning to tag from open vocabulary labels,” in Proc. ECML PKDD, Barcelona, Spain, 2010: Springer, pp. 211–226. [4] P. Lamere, “Social tagging and music information retrieval,” Journal of New Music Research, vol. 37, no. 2, pp. 101–114, 2008. [5] Y.-H. Yang, D. Bogdanov, P. Herrera, and M. Sordo, “Music retagging using label propagation and robust principal component analysis,” in Proc. WWW, Lyon, France, 2012: ACM, pp. 869–876. [6] J. Lee, J. Park, K. Kim, and J. Nam, “Samplecnn: End-to-end deep convolutional neural networks using very small filters for music classification,” Applied Sciences, vol. 8, no. 1, p. 150, 2018. [7] T. Wei and Y.-F. Li, “Does Tail Label Help for Large-Scale Multi-Label Learning,” IEEE Transactions on Neural Networks and Learning Systems, 2019. [8] X. Zhu and Z. Ghahramani, “Learning from labeled and unlabeled data with label propagation,” Tech. Rep. CMU-CALD-02–107, 2002. [9] C. Tang, X. Liu, P. Wang, C. Zhang, M. Li, and L. Wang, “Adaptive hypergraph embedded semi-supervised multi-label image annotation,” IEEE Transactions on Multimedia, vol. 21, no. 11, pp. 2837–2849, 2019. [10] D. Liu, S. Yan, X.-S. Hua, and H.-J. Zhang, “Image retagging using collaborative tag propagation,” IEEE Transactions on Multimedia, vol. 13, no. 4, pp. 702–712, 2011. [11] B. Frenay and M. Verleysen, “Classification in the presence of label noise: A survey,” IEEE Transactions on Neural Networks and Learning Systems, vol. 25, no. 5, pp. 845–869, 2014. [12] A. Bertoni, M. Frasca, and G. Valentini, “COSNet: a cost sensitive neural network for semi-supervised learning in graphs,” in Proc. ECML PKDD’11, Athens, Greece, 2011: Springer, pp. 219–234. [13] J. L. Leevy, T. M. Khoshgoftaar, R. A. Bauder, and N. Seliya, “A survey on addressing high-class imbalance in big data,” Journal of Big Data, vol. 5, no. 1, Jan. 2018. [14] G. Wu, Y. Tian, and D. Liu, “Cost-sensitive multi-label learning with positive and negative label pairwise correlations,” Neural Networks, vol. 108, pp. 411–423, 2018. [15] W. Fan, S. J. Stolfo, J. Zhang, and P. K. Chan, “AdaCost: misclassification cost-sensitive boosting,” in Proc. ICML, 1999, vol. 99, pp. 97–105. [16] Y.-H. Lin, C.-H. Chung, and H. H. Chen, “Playlist-Based Tag Propagation for Improving Music Auto-Tagging,” in Proc. EUSIPCO, Rome, Italy, 2018: IEEE, pp. 2270–2274. [17] M.-L. Zhang and Z.-H. Zhou, “ML-KNN: A lazy learning approach to multi-label learning,” Pattern Recognition, vol. 40, no. 7, pp. 2038–2048, 2007. [18] J. Salminen, V. Yoganathan, J. Corporan, B. J. Jansen, and S.-G. Jung, “Machine learning approach to auto-tagging online content for content marketing efficiency: A comparative analysis between methods and content type,” Journal of Business Research, vol. 101, pp. 203–217, 2019. [19] Y. Li, J. Liu, and J. Ren, “Social recommendation model based on user interaction in complex social networks,” PloS One, vol. 14, no. 7, Oct. 2019. [20] S. Baluja et al., “Video suggestion and discovery for youtube: taking random walks through the view graph,” in Proc. WWW, Beijin, China, 2008: ACM, pp. 895–904. [21] G. Feki, A. Ksibi, A. B. Ammar, and C. B. Amar, “Improving image search effectiveness by integrating contextual information,” in Proc. CBMI, Veszprém, Hungary, 2013: IEEE, pp. 149–154. [22] T. Patel and S. Gandhi, “A survey on context based similarity techniques for image retrieval,” in Proc. ICIMIA, Bangalore, India, 2017: IEEE, pp. 219–223. [23] P. Knees and M. Schedl, “A survey of music similarity and recommendation from music context data,” ACM Transactions on Multimedia Computing, Communications, and Applications, vol. 10, no. 1, p. 2, 2013. [24] C.-M. Chen, M.-F. Tsai, J.-Y. Liu, and Y.-H. Yang, “Music recommendation based on multiple contextual similarity information,” in Proc. WI-IAT, 2013: IEEE Computer Society, vol. 01, pp. 65–72. [25] J. L. Moore, S. Chen, T. Joachims, and D. Turnbull, “Learning to Embed Songs and Tags for Playlist Prediction,” in Proc. ISMIR, Porto, Portugal, 2012, vol. 12, pp. 349–354. [26] C.-H. Chung, Y. Chen, and H. H. Chen, “Exploiting Playlists for Representation of Songs and Words for Text-Based Music Retrieval,” in Proc. ISMIR, Suzhou, China, 2017, pp. 478–485. [27] K. Tsukuda, K. Ishida, and M. Goto, “Lyric Jumper: A Lyrics-Based Music Exploratory Web Service by Modeling Lyrics Generative Process,” in Proc. ISMIR, Suzhou, China, 2017, pp. 544–551. [28] J. H. Kim, B. Tomasik, and D. Turnbull, “Using Artist Similarity to Propagate Semantic Information,” in Proc. ISMIR, Kobe, Japan, 2009, vol. 9, pp. 375–380. [29] J. Wan and Y. Wang, “Cost-Sensitive Label Propagation for Semi-Supervised Face Recognition,” IEEE Transactions on Information Forensics and Security, vol. 14, no. 7, pp. 1729–1743, 2018. [30] S. Makki, R. Haque, Y. Taher, Z. Assaghir, M.-S. Hacid, and H. Zeineddine, “A Cost-Sensitive Cosine Similarity K-Nearest Neighbor for Credit Card Fraud Detection,” in Proc. BDCSIntell, Hadath, Lebanon, 2018, pp. 42–47. [31] C. Jin and S.-W. Jin, “Content-based image retrieval model based on cost sensitive learning,” Journal of Visual Communication and Image Representation, vol. 55, pp. 720–728, 2018. [32] N. Begwani, S. Harsola, R. Agrawal, “Learning from weights: a cost-sensitive approach for ad retrieval”, arXiv preprint arXiv:1811.12776, 2018. [33] H.-Y. Lo, J.-C. Wang, H.-M. Wang, and S.-D. Lin, “Cost-sensitive multi-label learning for audio tag annotation and retrieval,” IEEE Transactions on Multimedia, vol. 13, no. 3, pp. 518–529, 2011. | |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/21183 | - |
| dc.description.abstract | 音樂自動標籤的成效取決於訓練資料的品質。然而實際上，在人工標註的訓練資料中，歌曲與標籤的連結可能有誤（假陽性）或是有所遺漏（假陰性）。在此論文中，我們提出了「代價敏感的標籤傳遞學習法」，來改善自動標籤系統。首先，我們利用音樂周邊資訊來篩選出彼此相似的歌曲，並在它們之間傳遞標籤。然後，我們再把傳遞的標籤和原始的標籤一起用來最佳化自動標籤模型。另外，為了提升抗噪能力，我們把代價敏感的機制結合進模型的損失函數，以調整陽性連結相對於陰性連結的權重。接著，將所提出的方法用於訓練三個自動標籤模型，以檢測其成效。這三個模型分別為：CNN、CRNN和SampleCNN。其中，我們採用百萬歌曲資料集（Million Song Dataset）作為訓練資料，並使用四種不同的音樂周邊資訊：歌手、歌單、標籤以及聆聽者來衡量歌曲的相似性。實驗結果顯示：一、所提出的方法能夠成功的提升這三個模型的效能。二、代價敏感的損失函數有助於減少遺漏標籤的影響。三、歌手資訊比其他三種周邊資訊更適合做為標籤傳遞的媒介。 | zh_TW |
| dc.description.abstract | The performance of music auto-tagging depends on the quality of training data. In practice, the links between songs and tags in the manually labeled training data can be incorrect (false positive) or missing (false negative). In this paper, we propose a cost-sensitive tag propagation learning method to improve auto-tagging. Specifically, we exploit music context to determine similar songs and propagate tags between them. Both propagated tags and original tags are used to optimize the auto-tagging models, and cost-sensitivity is incorporated into the loss function to enhance the robustness by adjusting the weight of relevant (positive) links with respect to irrelevant (negative) links. The proposed method is tested on three auto-tagging models: 2D-CNN, CRNN, and SampleCNN. The Million Song Dataset is used for training, and four music contexts, artist, playlist, tag, and listener, are used for song similarity measurement. The experimental results show 1) The proposed method can successfully improve the performance of the three auto-tagging models, 2) The cost-sensitive loss function helps reduce the impact of missing tags, and 3) The artist music context is more powerful for tag propagation than the other three music contexts. | en |
| dc.description.provenance | Made available in DSpace on 2021-06-08T03:28:17Z (GMT). No. of bitstreams: 1 ntu-108-R04942055-1.pdf: 1332915 bytes, checksum: e033a1a519e02c9e39b0935e09f8dcc7 (MD5) Previous issue date: 2019 | en |
| dc.description.tableofcontents | 誌謝 ii
ABSTRACT iv CONTENTS v LIST OF FIGURES vii LIST OF TABLES viii Chapter 1 Introduction 1 Chapter 2 Related Work 5 2.1 Tag Propagation 5 2.2 Contextual Information 6 2.3 Cost-Sensitive Learning 6 Chapter 3 Methodology 8 3.1 Basic Training Process 9 3.2 Tag Propagation Mechanism 10 3.3 Cost-Sensitive Learning 13 Chapter 4 Experimental Setup 15 4.1 Datasets 15 4.2 Auto-Tagging Models 16 4.3 Evaluation Tasks 17 4.4 Baseline Training Processes 18 4.5 Parameter Tuning 18 Chapter 5 Result and Discussion 20 5.1 Performance Comparison 20 5.2 Effect of Tag Propagation 22 5.3 Effect of Cost-Sensitive Learning 26 Chapter 6 Conclusion 31 REFERENCES 33 APPENDIX I 37 APPENDIX II 38 | |
| dc.language.iso | en | |
| dc.subject | 音樂檢索系統 | zh_TW |
| dc.subject | 音樂自動標籤 | zh_TW |
| dc.subject | 標籤傳遞 | zh_TW |
| dc.subject | 音樂周邊資訊 | zh_TW |
| dc.subject | 代價敏感學習 | zh_TW |
| dc.subject | cost-sensitive learning | en |
| dc.subject | tag propagation | en |
| dc.subject | music information retrieval | en |
| dc.subject | music auto-tagging | en |
| dc.subject | music contextual information | en |
| dc.title | 基於音樂周邊資訊之音樂自動標籤系統優化 | zh_TW |
| dc.title | Improving Music Auto-Tagging System by Exploiting Music Context | en |
| dc.type | Thesis | |
| dc.date.schoolyear | 108-1 | |
| dc.description.degree | 碩士 | |
| dc.contributor.oralexamcommittee | 楊奕軒,蔡銘峰,陳怡安 | |
| dc.subject.keyword | 音樂自動標籤,音樂檢索系統,標籤傳遞,音樂周邊資訊,代價敏感學習, | zh_TW |
| dc.subject.keyword | music auto-tagging,music information retrieval,tag propagation,cost-sensitive learning,music contextual information, | en |
| dc.relation.page | 39 | |
| dc.identifier.doi | 10.6342/NTU201800616 | |
| dc.rights.note | 未授權 | |
| dc.date.accepted | 2018-08-17 | |
| dc.contributor.author-college | 電機資訊學院 | zh_TW |
| dc.contributor.author-dept | 電信工程學研究所 | zh_TW |
| 顯示於系所單位： | 電信工程學研究所 | |
文件中的檔案：
| 檔案 | 大小 | 格式 | |
|---|---|---|---|
| ntu-108-1.pdf 未授權公開取用 | 1.3 MB | Adobe PDF |
系統中的文件，除了特別指名其著作權條款之外，均受到著作權保護，並且保留所有的權利。