Please use this identifier to cite or link to this item:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/56998
Full metadata record
???org.dspace.app.webui.jsptag.ItemTag.dcfield??? | Value | Language |
---|---|---|
dc.contributor.advisor | 鄭卜壬(Pu-Jen Cheng) | |
dc.contributor.author | Cheng-Hsuan Tsai | en |
dc.contributor.author | 蔡誠軒 | zh_TW |
dc.date.accessioned | 2021-06-16T06:32:39Z | - |
dc.date.available | 2019-09-04 | |
dc.date.copyright | 2014-09-04 | |
dc.date.issued | 2014 | |
dc.date.submitted | 2014-08-05 | |
dc.identifier.citation | [1] Natalie Aizenberg, Yehuda Koren, and Oren Somekh. Build your own music recommender by modeling internet radio streams. In WWW, pages 1–10, 2012.
[2] Shuo Chen, Joshua L. Moore, Douglas Turnbull, and Thorsten Joachims. Playlist prediction via metric embedding. In KDD, pages 714–722, 2012. [3] Shuo Chen, Jiexun Xu, and Thorsten Joachims. Multi-space probabilistic sequence modeling. In KDD, pages 865–873, 2013. [4] Ofer Dekel, Ran Gilad-Bachrach, Ohad Shamir, and Lin Xiao. Optimal distributed online prediction using mini-batches. Journal of Machine Learning Research, 13: 165–202, 2012. [5] George Karypis and Vipin Kumar. Multilevel k-way partitioning scheme for irregular graphs. J. Parallel Distrib. Comput., 48(1):96–129, 1998. [6] Vlado Keselj. Speech and language processing (second edition) daniel jurafsky and james h. martin (stanford university and university of colorado at boulder) pearson prentice hall, 2009, hardbound, isbn 978-0-13-187321-6. Computational Linguistics, 35(3):463–466, 2009. [7] Yehuda Koren, Robert M. Bell, and Chris Volinsky. Matrix factorization techniques for recommender systems. IEEE Computer, 42(8):30–37, 2009. [8] François Maillet, Douglas Eck, Guillaume Desjardins, and Paul Lamere. Steerable playlist generation by learning song similarity from radio station playlists. In ISMIR, pages 345–350, 2009. [9] Yariv Maron, Michael Lamar, and Elie Bienenstock. Sphere embedding: An application to part-of-speech induction. In NIPS, pages 1567–1575, 2010. [10] Brian McFee and Gert R. G. Lanckriet. The natural language of playlists. In ISMIR, pages 537–542, 2011. [11] Robert Ragno, Christopher J. C. Burges, and Cormac Herley. Inferring similarity between music objects with application to playlist generation. In Multimedia Information Retrieval, pages 73–80, 2005. [12] Benjamin Recht, Christopher Re, Stephen J. Wright, and Feng Niu. Hogwild: A lock-free approach to parallelizing stochastic gradient descent. In NIPS, pages 693–701, 2011. [13] Steffen Rendle, Christoph Freudenthaler, and Lars Schmidt-Thieme. Factorizing personalized markov chains for next-basket recommendation. In WWW, pages 811–820, 2010. [14] Martin Zinkevich, Markus Weimer, Alexander J. Smola, and Lihong Li. Parallelized stochastic gradient descent. In NIPS, pages 2595–2603, 2010. | |
dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/56998 | - |
dc.description.abstract | 馬可夫空間(Logistic Markov Embedding)已被驗證為能夠有效學習物件序列模型的演算法。
藉由此演算法,使用者可以從大量的歷史物件序列中,產生類似的新物件序列,藉以達成序列推薦的應用。 但是,由於計算馬可夫空間使用的演算法具有很高的時間複雜度,此演算法幾乎無法應用到實際擁有大量資料的系統之中。 為了增加其實用性與可擴張的規模,已有少數研究成果試著對此演算法進行加速,將原本的物件集合切割成許多小型的集合,並對每個小集合計算獨立的馬可夫空間。 在此論文中,我們提出了一種新的集合分群方式,允許分群後的小型集合彼此在必要的部份重疊。 我們驗證了新的分群方式相對於上述目前最好的加速演算法可以達到更高的準確度,並且可以在相同或是更短的運算時間內完成。 | zh_TW |
dc.description.abstract | Logistic Markov Embedding (LME) has become a popular branch on the research of sequential item recommendation, such as music playlist generation.
But, since LME is an algorithm with very high time complexity, it has a poor scalability and is not able to carry a huge dataset with many items. Hence, several approaches are trying to decrease the time complexity of LME, while keeping the prediction accuracy. In this paper, we present a new speed-up approach for LME, which convert the original item set into several smaller and overlapped clusters, then train a LME for each cluster. We show that this new clustering algorithm is able to get a better performance in a shorter training time compared to the current best speed-up approach. | en |
dc.description.provenance | Made available in DSpace on 2021-06-16T06:32:39Z (GMT). No. of bitstreams: 1 ntu-103-R00922005-1.pdf: 375293 bytes, checksum: b0feeff85f0be85a6605b9ae205f56a1 (MD5) Previous issue date: 2014 | en |
dc.description.tableofcontents | 誌謝 ii
摘要 iii Abstract iv 1 Introduction 1 2 Related Works 3 3 Prerequisites 5 3.1 Learning process of Logistic Markov Embedding 6 3.2 Multi-LME 7 4 Overlapped-LME 10 4.1 Design the transition probability of Overlapped-LME 11 4.2 Clustering algorithms 13 4.3 Time complexity 14 5 Experiment 16 5.1 Overlapped-LME v.s. Multi-LME 17 5.2 Effect of two-steps probability 20 5.3 Effect of different clustering algorithms 20 6 Conclusions & Future work 22 Bibliography 23 | |
dc.language.iso | en | |
dc.title | 使用重疊馬可夫空間計算物件序列模型 | zh_TW |
dc.title | Modeling Item Sequences by Overlapped Markov Embeddings | en |
dc.type | Thesis | |
dc.date.schoolyear | 102-2 | |
dc.description.degree | 碩士 | |
dc.contributor.oralexamcommittee | 盧文祥,蔡宗翰(Richard Tzong-Han Tsai),邱志義(Chih-Yi Chiu) | |
dc.subject.keyword | 馬可夫空間,推薦系統,物件序列,加速演算法,機器學習, | zh_TW |
dc.subject.keyword | Recommendation System,Markov Embedding,Speed-up Algorithm,Machine Learning,Parallel Computing,Item Sequences, | en |
dc.relation.page | 24 | |
dc.rights.note | 有償授權 | |
dc.date.accepted | 2014-08-05 | |
dc.contributor.author-college | 電機資訊學院 | zh_TW |
dc.contributor.author-dept | 資訊工程學研究所 | zh_TW |
Appears in Collections: | 資訊工程學系 |
Files in This Item:
File | Size | Format | |
---|---|---|---|
ntu-103-1.pdf Restricted Access | 366.5 kB | Adobe PDF |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.