過去幾年中，數位音樂在網路所佔的比例有明顯成長，許多音樂平台及網站如雨後春筍般在網路的世界裡興起，這些音樂平台擁有大量的音樂相關資訊，雖可增加資訊豐富度，卻不容易讓使用者輕易獲取與自身確切相關的資訊，因此了解使用者的喜好變成音樂平台能超越競爭對手的決定性優勢。
推薦系統的價值在於可給予被推薦者合適的建議，利用特定的資訊過濾技術，幫助使用者從大量的資料中選出可能會有興趣的主題或資源。而在大眾分類法與資源共享系統下，使用者可利用標籤收集有興趣的資源，因標籤是使用者自行定義的，被認為可明確代表使用者意見與喜好，因此有學者將標籤概念加入推薦系統中，改善協同過濾推薦下的評價差異性等問題，然而在音樂社交平台上還包含更多隱性資訊如用戶的收聽行為、朋友圈等，都是可考量的因素。
本研究目的在於提出一個考慮用戶之標籤集、歌曲播放次數與朋友關係的相似度函數，由於用戶對於音樂方面的原始標籤並無統一用法，為了足以表示對音樂上的興趣相似，先將網站上用戶所註記的音樂類型標籤先用語意網其標籤集將其標準化後，對用戶做分群，針對目標用戶，先將其歸類到特定群組後，再與群內用戶做歌曲喜好的相似度分析，找出推薦歌曲列表；除了以標籤代表個人偏好來計算相似度之外，並將用戶的歌曲播放次數、歌曲的標籤次數等隱性評價作為權重，而實驗結果發現，利用標準化後的用戶標籤權重計算推薦效果的平均準度均值(MAP)為4.6%，比以播放次數為權重的2.4%較好，朋友關係加入於用戶標籤中，亦能改善推薦效果至5.1%，加入分群技術後也有助於推薦效果至5.7%。

Past years, many music platforms spring up in the Internet world. To help users obtain precise information relevant to themselves from abundant music-related information, understanding users’ preferences have become the decisive advantage to exceed other competitors. With the help of folksonomy and social resource sharing systems, more platforms provide users with tags to collect resources of interest. Users’ tag sets can be regarded as their preferences since tags are user-defined. Research on applying tags to recommender systems has been extensively done. However, more hidden information still can be considered.
This study propose a similarity function utilizing user’s tag sets, play counts of songs and friendship in order to recommend effectively. Since users don’t have uniform usage of tags, we standardize users’ tags for music to indicate the users’ interest precisely. After clustering users using the standardized tags, we calculate the interest similarity of music type between two users. The weights of this similarity function consist of user’s implicit evaluation including tags represented his preference, listening frequency and tag frequency of each song and friendship, etc. Experiment result shows that MAP of using standardized tag-weighted similarity is 4.6%, which is better than 2.4% of play-count-weighted similarity. Friendship can improve MAP to 5.1%, and clustering also help MAP increase to 5.7%.

Cantador, I., Konstas, I., & Jose, J. M. (2011). Categorising social tags to improve folksonomy-based recommendations. Web Semantics: Science, Services and Agents on the World Wide Web, 9(1), 1-15. doi: http://dx.doi.org/10.1016/j.websem.2010.10.001
Chen, Y.-L., & Chiu, Y.-T. (2011). An IPC-based vector space model for patent retrieval. Information Processing & Management, 47(3), 309-322. doi: http://dx.doi.org/10.1016/j.ipm.2010.06.001
Chiang, H.-S., & Huang, T.-C. (2015). User-adapted travel planning system for personalized schedule recommendation. Information Fusion, 21(0), 3-17. doi: http://dx.doi.org/10.1016/j.inffus.2013.05.011
Choi, K., & Suh, Y. (2013). A new similarity function for selecting neighbors for each target item in collaborative filtering. Knowledge-Based Systems, 37(0), 146-153. doi: http://dx.doi.org/10.1016/j.knosys.2012.07.019
Goldberg, D., Nichols, D., Oki, B. M., & Terry, D. (1992). Using collaborative filtering to weave an information tapestry. Commun. ACM, 35(12), 61-70. doi: 10.1145/138859.138867
Huang, C.-L., Yeh, P.-H., Lin, C.-W., & Wu, D.-C. (2014). Utilizing user tag-based interests in recommender systems for social resource sharing websites. Knowledge-Based Systems, 56(0), 86-96. doi: http://dx.doi.org/10.1016/j.knosys.2013.11.001
Kim, H. H. (2013). A semantically enhanced tag-based music recommendation using emotion ontology. Paper presented at the Proceedings of the 5th Asian conference on Intelligent Information and Database Systems - Volume Part II, Kuala Lumpur, Malaysia.
Krestel, R., & Fankhauser, P. (2012). Personalized topic-based tag recommendation. Neurocomputing, 76(1), 61-70. doi: http://dx.doi.org/10.1016/j.neucom.2011.04.034
Li, Y.-M., Hsiao, H.-W., & Lee, Y.-L. (2013). Recommending social network applications via social filtering mechanisms. Information Sciences, 239(0), 18-30. doi: http://dx.doi.org/10.1016/j.ins.2013.03.041
Li, Y.-M., Wu, C.-T., & Lai, C.-Y. (2013). A social recommender mechanism for e-commerce: Combining similarity, trust, and relationship. Decision Support Systems, 55(3), 740-752. doi: http://dx.doi.org/10.1016/j.dss.2013.02.009
Liang, T.-P., Yang, Y.-F., Chen, D.-N., & Ku, Y.-C. (2008). A semantic-expansion approach to personalized knowledge recommendation. Decision Support Systems, 45(3), 401-412. doi: http://dx.doi.org/10.1016/j.dss.2007.05.004
Mahdavi, M., Chehreghani, M. H., Abolhassani, H., & Forsati, R. (2008). Novel meta-heuristic algorithms for clustering web documents. Applied Mathematics and Computation, 201(1–2), 441-451. doi: http://dx.doi.org/10.1016/j.amc.2007.12.058
MBA智庫百科. (2014). 網路音樂的市場發展趨勢. from http://wiki.mbalib.com/zh-tw/%E7%BD%91%E7%BB%9C%E9%9F%B3%E4%B9%90
Rui, X., & Wunsch, D., II. (2005). Survey of clustering algorithms. Neural Networks, IEEE Transactions on, 16(3), 645-678. doi: 10.1109/TNN.2005.845141
Salton, G., Wong, A., & Yang, C. S. (1975). A vector space model for automatic indexing. Commun. ACM, 18(11), 613-620. doi: 10.1145/361219.361220
Schedl, M., Widmer, G., Knees, P., & Pohle, T. (2011). A music information system automatically generated via Web content mining techniques. Information Processing & Management, 47(3), 426-439. doi: http://dx.doi.org/10.1016/j.ipm.2010.09.002
StrategyAnalytics. (2013). Global Digital Music Gains Struggle to Offset Declining Physical Music Sales. from http://www.strategyanalytics.com/default.aspx?mod=pressreleaseviewer&a0=5456
Su, J.-H., Chang, W.-Y., & Tseng, V. S. (2013). Personalized Music Recommendation by Mining Social Media Tags. Procedia Computer Science, 22(0), 303-312. doi: http://dx.doi.org/10.1016/j.procs.2013.09.107
Tatli, I., & Birturk, A. (2011, 11-11 Dec. 2011). A Tag-Based Hybrid Music Recommendation System Using Semantic Relations and Multi-domain Information. Paper presented at the Data Mining Workshops (ICDMW), 2011 IEEE 11th International Conference on.
Zhang, L., Hu, C., Chen, Q., Chen, Y., & Shi, Y. (2012). Domain Knowledge Based Personalized Recommendation Model and Its Application in Cross-selling. Procedia Computer Science, 9(0), 1314-1323. doi: http://dx.doi.org/10.1016/j.procs.2012.04.144
Zhang, Z.-K., Yu, L., Fang, K., You, Z.-Q., Liu, C., Liu, H., & Yan, X.-Y. (2014). Website-oriented recommendation based on heat spreading and tag-aware collaborative filtering. Physica A: Statistical Mechanics and its Applications, 399(0), 82-88. doi: http://dx.doi.org/10.1016/j.physa.2013.12.030
林宜瑩. (2010). 利用時間因子與名詞片語之文獻主題追蹤法. 國立成功大學.
創市際市場研究顧問. (2011). 2011年05月 創市際音樂下載篇. from http://www.insightxplorer.com/specialtopic/2011_05_13.htm