研究生: |
許立璇 Li-Hsuan Hsu |
---|---|
論文名稱: |
以情緒與場景為基礎的英文音樂推薦系統 English Music Recommendation System Based on Emotion and Scene |
指導教授: |
徐俊傑
Chiun-chieh Hsu |
口試委員: |
王有禮
Yue-li Wang 洪政煌 Cheng-Huang Hung |
學位類別: |
碩士 Master |
系所名稱: |
管理學院 - 資訊管理系 Department of Information Management |
論文出版年: | 2021 |
畢業學年度: | 109 |
語文別: | 中文 |
論文頁數: | 56 |
中文關鍵詞: | 推薦系統 、音樂推薦 、特徵擷取 、多標籤分類器 |
外文關鍵詞: | Recommendation system, Music Recommender, Feature Extraction, Multi-label Classification |
相關次數: | 點閱:208 下載:0 |
分享至: |
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
由於網際網路的蓬勃發展,串流音樂市場也隨之成長,使我們可以隨時取得
各種音樂。然而在眾多的歌曲中,要如何讓使用者以較低的搜尋成本獲取有興趣
的歌曲即成為一個困難的挑戰,音樂推薦系統即擔任了重要的角色。
本研究提出一個分析歌詞情緒與場景的英文歌曲推薦系統,利用ConceptNet
語義網絡與NRC 情緒詞庫分別抽取歌詞的情緒詞與場景詞。首先將歌曲視作多
種情緒主題的分布,利用隱含狄利克雷分布模型取得情緒主題機率分布作為歌曲
的情緒特徵。且為了理解歌詞中常有的心情轉折,提出以歌詞結構為基礎的標記
策略,將重要的副歌段落與其他段落分別訓練。同時利用TF-IDF 詞權重技術獲
得歌詞的場景特徵,再以多標籤分類器找到有相同場景的候選歌曲。最後計算情
緒特徵與場景特徵的相似性,找到目標歌曲的歌曲推薦清單。
經由實驗發現,本研究提出的歌曲推薦系統相較目前方法,在推薦不同歌曲
數下,系統合適度評價的精確率可提高6.1~20%,而在使用者系統偏好的實驗中
能獲得高於23%的偏好比例。
Due to the rapid growth of the Internet, the streaming music market has gradually
grown. This trend allows us to obtain plenty of songs at any time. Nevertheless, among
the huge amount, how to let users obtain the song they are interested in at a lower search
cost has become a difficult challenge. The music recommendation system has played
an important role in this issue.
In this thesis, we propose an English music recommendation system that analyzes
the emotion and scene context of lyrics. The ConceptNet semantic network and the
NRC emotional vocabulary are used to extract the emotion terms and scene terms of
the lyrics respectively. We regard song as a distribution of multiple emotional themes
and train Labeled LDA model to obtain the emotion distribution as the emotion feature.
Furthermore, in order to better understand the mood transitions often found in lyrics, a
label strategy based on the lyrics structure is proposed to train chorus and other
paragraphs separately. Meanwhile, a song uses TF-IDF word weighting technology to
obtain the scene feature of the lyric, and then make use of the multi-label classifier to
find candidate songs with the same scene. Finally, calculate the similarity between the
emotion features and the scene features, and find the song recommendation list of the
target song.
Through experiments, under different number of songs recommended, it is found
that our method can improve the accuracy of the system suitability evaluation by 6.1-
20% compared with the current method. As well as the experiment of user system
preference can be higher than 23%.
[1] S. Baccianella, A. Esuli, and F. Sebastiani, “SentiWordNet 3.0: An Enhanced
Lexical Resource for Sentiment Analysis and Opinion Mining,” in Proceedings
of the Seventh International Conference on Language Resources and
Evaluation, pp. 2200-2204, 2010.
[2] D. Blei, A. Ng, and M. Jordan, “Latent Dirichlet Allocation,” Journal of
Machine Learning Research, vol. 3, no. 7, pp. 993-1022, 2003.
[3] R. Cai, C. Zhang, C. Wang, L. Zhang, and W. Y. Ma, “MusicSense: contextual
music recommendation using emotional allocation modeling,” in Proceedings
of the 15th International Conference on Multimedia, pp. 553-556, 2007.
[4] C. M. Chen, M. F. Tsai, J. Y. Liu, and Y. H. Yang, “Using emotional context
from article for contextual music recommendation,” in ACM international
conference on Multimedia, pp. 649–652, 2013.
[5] K. Chen, S. L. Huang, Y. Y. Shih, and Y. J. Chen, “Extended-HowNet: A
Representational Framework for Concepts,” in International Joint Conference
on Natural Language Processing, pp. 1-6, 2005.
[6] J. Choi, J. Song, and Y. Kim, “An Analysis of Music Lyrics by Measuring the
Distance of Emotion and Sentiment,” in International Conference on Software
Engineering, Artificial Intelligence, Networking and Parallel/Distributed
Computing (SNPD), pp. 176-181, 2018.
[7] T. Cover, and P. Hart, “Nearest neighbor pattern classification,” IEEE
Transactions on Information Theory, vol. 13, no. 1, pp. 21-27, Jan. 1967.
[8] S. Deerwester, S. Dumais, G. Furnas, T. Landauer, and R. Harshman, “Indexing
by Latent Semantic Analysis,” Journal of the Association for Information
Science and Technology, vol. 41, no. 6, pp. 391-407, 1990.
[9] M. Fell, and C. Sporleder, “Lyrics-based Analysis and Classification of Music,”
in International Conference on Computational Linguistics, pp. 620–631, 2014.
[10] M. Furuya, H.-H. Huang, and K. Kawagoe, “Music classification method based
on lyrics for music therapy,” in International Database Engineering &
Applications Symposium, pp. 382–383, 2014.
[11] J. Gao, H. Yuan, L. Wang, and Y. Qian, “Evaluating Sentiment Similarity of
Songs Based on Social Media Data,” in International Conference on Service
Systems and Service Management, pp. 1-6, 2018.
48
[12] M. A. Hearst, S. T. Dumais, E. Osuna, J. Platt, and B. Scholkopf, “Support
vector machines,” IEEE Intelligent Systems and their Applications, vol. 13, no.
4, pp. 18-28, 1998.
[13] R. Hossain, M. R. K. R. Sarker, M. Mimo, A. A. Marouf, and B. Pandey,
“Recommendation Approach of English Songs Title Based on Latent Dirichlet
Allocation applied on Lyrics,” in International Conference on Electrical,
Computer and Communication Technologies, pp. 1-4, 2019.
[14] X. Hu, and J. S. Downie, “When Lyrics Outperform Audio for Music Mood
Classification: A Feature Analysis,” in International Society for Music
Information Retrieval Conference, pp. 619-624, 2010.
[15] P. Juslin, and P. Laukka, “Expression, Perception, and Induction of Musical
Emotions: A Review and a Questionnaire Study of Everyday Listening,”
Journal of New Music Research, vol. 33, no. 3, pp. 217 - 238, 2010.
[16] Y. E. Kim, E. M. Schmidt, R. Migneco, B. G. Morton, P. Richardson, J. J. Scott,
J. A. Speck, and D. Turnbull, “Music emotion recognition: A state of the art
review,” in International Society for Music Information and Retrieval
Conference, pp. 255-266, 2010.
[17] J. H. Lee, and J. S. Downie, “Survey Of Music Information Needs, Uses, And
Seeking Behaviours: Preliminary Findings,” in International Conference on
Music Information Retrieval, pp. 1-4, 2004.
[18] B. Logan, A. Kositsky, and P. Moreno, “Semantic analysis of song lyrics,” in
IEEE International Conference on Multimedia and Expo, pp. 827-830, 2004.
[19] L. Lu, D. Liu, and H. Zhang, “Automatic mood detection and tracking of music
audio signals,” IEEE Transactions on Audio, Speech, and Language Processing,
vol. 14, no. 1, pp. 5-18, 2006.
[20] R. Mayer, R. Neumayer, and A. Rauber, “Rhyme and Style Features for Musical
Genre Classification by Song Lyrics,” in International Conference on Music
Information Retrieval, pp. 337–342, 2008.
[21] R. Mayer, R. Neumayer, and A. Rauber, “Combination of audio and lyrics
features for genre classification in digital audio collections,” in International
conference on Multimedia, pp. 159-168, 2008.
[22] S. M. Mohammad, and P. D. Turney, “Crowdsourcing a Word-Emotion
Association Lexicon,” Computational Intelligence, vol. 29, no. 3, pp. 436–465,
49
2013.
[23] A. Navada, A. N. Ansari, S. Patil, and B. A. Sonkamble, “Overview of use of
decision tree algorithms in machine learning,” in IEEE Control and System
Graduate Research Colloquium, pp. 37-42, 2011.
[24] A. M. Oudenne, and S. E. Chasins, “Identifying the Emotional Polarity of Song
Lyrics through Natural Language Processing,” Berkeley, Department of
Electrical Engineering and Computer Sciences, 2010.
[25] H. Piliang, and R. Kusumaningrum, “Music Emotion Classification Based on
Indonesian Song Lyrics Using Recurrent Neural Network,” in International
Conference on Informatics and Computational Sciences, pp. 1-4, 2019.
[26] D. Ramage, D. Hall, R. Nallapati, and C. Manning, “Labeled LDA,” in
Conference on Empirical Methods in Natural Language Processing, pp. 248–
256, 2009.
[27] R. Rosa, D. Z. RodrÌguez, and G. Bressan, “Music recommendation system
based on user's sentiments extracted from social networks,” IEEE Transactions
on Consumer Electronics, vol. 61, no. 3, pp. 359-367, 2015.
[28] S. Sasaki, K. Yoshii, T. Nakano, M. Goto, and S. Morishima, “LyricsRadar: A
Lyrics Retrieval System Based on Latent Topics of Lyrics,” in International
Society for Music Information Retrieval Conference, pp. 585-590, 2014.
[29] H. Sharma, S. Gupta, Y. Sharma, and A. Purwar, “A New Model for Emotion
Prediction in Music,” in International Conference on Signal Processing and
Communication, pp. 156-161, 2020.
[30] K. Siriket, V. Sa-ing, and S. Khonthapagdee, “Mood classification from Song
Lyric using Machine Learning,” in 2021 9th International Electrical
Engineering Congress, pp. 476-478, 2021.
[31] R. Speer, J. Chin, and C. Havasi, “ConceptNet 5.5: An Open Multilingual Graph
of General Knowledge,” in AAAI Conference on Artificial Intelligence, pp.
4444–4451, 2019.
[32] StudioPros. "Songwriting Tip: Lyrical Themes.,"
https://studiopros.com/songwriting-tip-lyrical-themes/.
[33] G. Tzanetakis, and P. Cook, “Musical genre classification of audio signals,”
IEEE Trans. Speech Audio Process, vol. 10, no. 5, pp. 293-302, 2002.
[34] C.-C. Wang, “Using Emotion and Scene Features of Lyrics for Social Article
50
Music Recommendation,” National Cheng Kung University, College of
Electrical Engineering & Computer Science, 2017.
[35] A. B. Warriner, V. Kuperman, and M. Brysbaert, “Norms of valence, arousal,
and dominance for 13,915 English lemmas,” Behavior Research Methods, vol.
45, no. 4, pp. 1191-1207, 2013.