Mercurial > hg > hybrid-music-recommender-using-content-based-and-social-information

annotate Report/references.bib @ 27:ae650489d3a8

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Updated report
author Paulo Chiliguano <p.e.chiilguano@se14.qmul.ac.uk>
date Sun, 30 Aug 2015 15:49:27 +0100
parents e4bcfe00abf4
children a95e656907c3
rev   line source
p@27 1 @online{1_lecun_2015,
p@27 2 author={LeCun, Yann},
p@27 3 title={MNIST Demos on Yann LeCun's website},
p@27 4 url={http://yann.lecun.com/exdb/lenet/},
p@27 5 urldate={2015-8-30},
p@27 6 journal={Yann.lecun.com},
p@27 7 year={2015}
p@27 8 }
p@27 9 @online{1_brown_2014,
p@27 10 author={Brown, Larry},
p@27 11 title={Accelerate Machine Learning with the cuDNN Deep Neural Network Library},
p@27 12 url={http://devblogs.nvidia.com/parallelforall/accelerate-machine-learning-cudnn-deep-neural-network-library/},
p@27 13 urldate={2015-8-30},
p@27 14 journal={Parallel Forall},
p@27 15 year={2014}
p@27 16 }
p@27 17
p@27 18 @article{hinton2012deep,
p@27 19 title={Deep neural networks for acoustic modeling in speech recognition: The shared views of four research groups},
p@27 20 author={Hinton, Geoffrey and Deng, Li and Yu, Dong and Dahl, George E and Mohamed, Abdel-rahman and Jaitly, Navdeep and Senior, Andrew and Vanhoucke, Vincent and Nguyen, Patrick and Sainath, Tara N and others},
p@27 21 journal={Signal Processing Magazine, IEEE},
p@27 22 volume={29},
p@27 23 number={6},
p@27 24 pages={82--97},
p@27 25 year={2012},
p@27 26 publisher={IEEE}
p@27 27 }
p@27 28 @article{weston2012latent,
p@27 29 title={Latent collaborative retrieval},
p@27 30 author={Weston, Jason and Wang, Chong and Weiss, Ron and Berenzweig, Adam},
p@27 31 journal={arXiv preprint arXiv:1206.4603},
p@27 32 year={2012}
p@27 33 }
p@27 34
p@27 35 @MISC{Bastien-Theano-2012,
p@27 36 author = {Bastien, Fr{\'{e}}d{\'{e}}ric and Lamblin, Pascal and Pascanu, Razvan and Bergstra, James and Goodfellow, Ian J. and Bergeron, Arnaud and Bouchard, Nicolas and Bengio, Yoshua},
p@27 37 title = {Theano: new features and speed improvements},
p@27 38 year = {2012},
p@27 39 howpublished = {Deep Learning and Unsupervised Feature Learning NIPS 2012 Workshop},
p@27 40 abstract = {Theano is a linear algebra compiler that optimizes a user’s symbolically-specified
p@27 41 mathematical computations to produce efficient low-level implementations. In
p@27 42 this paper, we present new features and efficiency improvements to Theano, and
p@27 43 benchmarks demonstrating Theano’s performance relative to Torch7, a recently
p@27 44 introduced machine learning library, and to RNNLM, a C++ library targeted at
p@27 45 recurrent neural networks.}
p@27 46 }
p@27 47
p@27 48 @INPROCEEDINGS{bergstra+al:2010-scipy,
p@27 49 author = {Bergstra, James and Breuleux, Olivier and Bastien, Fr{\'{e}}d{\'{e}}ric and Lamblin, Pascal and Pascanu, Razvan and Desjardins, Guillaume and Turian, Joseph and Warde-Farley, David and Bengio, Yoshua},
p@27 50 month = jun,
p@27 51 title = {Theano: a {CPU} and {GPU} Math Expression Compiler},
p@27 52 booktitle = {Proceedings of the Python for Scientific Computing Conference ({SciPy})},
p@27 53 year = {2010},
p@27 54 location = {Austin, TX},
p@27 55 note = {Oral Presentation},
p@27 56 abstract = {Theano is a compiler for mathematical expressions in Python that combines the convenience of NumPy’s syntax with the speed of optimized native machine language. The user composes mathematical expressions in a high-level description that mimics NumPy’s syntax and semantics, while being statically typed and
p@27 57 functional (as opposed to imperative). These expressions allow Theano to provide symbolic differentiation. Before performing computation, Theano optimizes the choice of expressions, translates
p@27 58 them into C++ (or CUDA for GPU), compiles them into dynamically loaded Python modules, all automatically. Common machine learning algorithms implemented with Theano are from 1.6× to 7.5× faster than competitive alternatives (including those implemented with C/C++, NumPy/SciPy and MATLAB) when compiled for the
p@27 59 CPU and between 6.5× and 44× faster when compiled for the GPU. This paper illustrates how to use Theano, outlines the scope of the compiler, provides benchmarks on both CPU and GPU processors, and explains its overall design.}
p@27 60 }
p@27 61
p@27 62 @article{kereliuk15,
p@27 63 title={Deep Learning and Music Adversaries},
p@27 64 author={Kereliuk, Corey and Sturm, Bob L and Larsen, Jan},
p@27 65 journal={arXiv preprint arXiv:1507.04761},
p@27 66 year={2015}
p@27 67 }
p@27 68
p@27 69 @article{DBLP:journals/corr/abs-1305-1114,
p@27 70 author = {Djallel Bouneffouf},
p@27 71 title = {Towards User Profile Modelling in Recommender System},
p@27 72 journal = {CoRR},
p@27 73 volume = {abs/1305.1114},
p@27 74 year = {2013},
p@27 75 url = {http://arxiv.org/abs/1305.1114},
p@27 76 timestamp = {Sun, 02 Jun 2013 20:48:21 +0200},
p@27 77 biburl = {http://dblp.uni-trier.de/rec/bib/journals/corr/abs-1305-1114},
p@27 78 bibsource = {dblp computer science bibliography, http://dblp.org}
p@27 79 }
p@27 80
p@27 81 @ARTICLE{Yao2015453,
p@27 82 author={Yao, L. and Sheng, Q.Z. and Ngu, A.H.H. and Yu, J. and Segev, A.},
p@27 83 title={Unified collaborative and content-based web service recommendation},
p@27 84 journal={IEEE Transactions on Services Computing},
p@27 85 year={2015},
p@27 86 volume={8},
p@27 87 number={3},
p@27 88 pages={453-466},
p@27 89 doi={10.1109/TSC.2014.2355842},
p@27 90 art_number={6894179},
p@27 91 note={},
p@27 92 url={http://www.scopus.com/inward/record.url?eid=2-s2.0-84932619562&partnerID=40&md5=4483a697e12fc53f620393586f85aebe},
p@27 93 document_type={Article},
p@27 94 source={Scopus},
p@27 95 }
p@27 96 @online{1_blogseagatesoftcom_2015,
p@27 97 author={Blog.seagatesoft.com,},
p@27 98 title={Belajar Sistem Perekomendasi « Corat-coret di Halaman Web},
p@27 99 url={http://blog.seagatesoft.com/2013/07/14/belajar-sistem-perekomendasi/},
p@27 100 urldate={2015-8-29},
p@27 101 year={2015}
p@27 102 }
p@27 103
p@27 104 @inproceedings{sarwar2001item,
p@27 105 title={Item-based collaborative filtering recommendation algorithms},
p@27 106 author={Sarwar, Badrul and Karypis, George and Konstan, Joseph and Riedl, John},
p@27 107 booktitle={Proceedings of the 10th international conference on World Wide Web},
p@27 108 pages={285--295},
p@27 109 year={2001},
p@27 110 organization={ACM}
p@27 111 }
p@27 112
p@27 113 @CONFERENCE{Hu2008263,
p@27 114 author={Hu, Y. and Volinsky, C. and Koren, Y.},
p@27 115 title={Collaborative filtering for implicit feedback datasets},
p@27 116 journal={Proceedings - IEEE International Conference on Data Mining, ICDM},
p@27 117 year={2008},
p@27 118 pages={263-272},
p@27 119 doi={10.1109/ICDM.2008.22},
p@27 120 art_number={4781121},
p@27 121 note={},
p@27 122 url={http://www.scopus.com/inward/record.url?eid=2-s2.0-67049164166&partnerID=40&md5=01238b08208962fd0fdcc7503fa3af99},
p@27 123 document_type={Conference Paper},
p@27 124 source={Scopus},
p@27 125 }
p@27 126
p@27 127 @online{2_the_economist_2005,
p@27 128 author={The Economist,},
p@27 129 title={United we find},
p@27 130 url={http://www.economist.com/node/3714044},
p@27 131 urldate={2015-8-28},
p@27 132 year={2005}
p@27 133 }
p@27 134
p@27 135 @online{1_siddharths_blog_2013,
p@27 136 author={},
p@27 137 title={Recommendation Engine},
p@27 138 url={https://spatnaik77.wordpress.com/2013/07/17/recommendation-engine/},
p@27 139 urldate={2015-8-28},
p@27 140 year={2013}
p@27 141 }
p@27 142
p@27 143 @online{1_deeplearning.net_2015,
p@27 144 author={Deeplearning.net,},
p@27 145 title={Convolutional Neural Networks (LeNet) — DeepLearning 0.1 documentation},
p@27 146 url={http://deeplearning.net/tutorial/lenet.html},
p@27 147 urldate={2015-8-28},
p@27 148 year={2015}
p@27 149 }
p@27 150 @ARTICLE{Casey2008668,
p@27 151 author={Casey, M.A. and Veltkamp, R. and Goto, M. and Leman, M. and Rhodes, C. and Slaney, M.},
p@27 152 title={Content-based music information retrieval: Current directions and future challenges},
p@27 153 journal={Proceedings of the IEEE},
p@27 154 year={2008},
p@27 155 volume={96},
p@27 156 number={4},
p@27 157 pages={668-696},
p@27 158 doi={10.1109/JPROC.2008.916370},
p@27 159 art_number={4472077},
p@27 160 note={},
p@27 161 url={http://www.scopus.com/inward/record.url?eid=2-s2.0-64649105397&partnerID=40&md5=2d8ec7231e10bc686566dd419ce47ae8},
p@27 162 document_type={Article},
p@27 163 source={Scopus},
p@27 164 }
p@27 165
p@27 166 @CONFERENCE{Celma2006,
p@27 167 author={Celma, O. and Herrera, P. and Serra, X.},
p@27 168 title={Bridging the music semantic gap},
p@27 169 journal={CEUR Workshop Proceedings},
p@27 170 year={2006},
p@27 171 volume={187},
p@27 172 page_count={12},
p@27 173 note={},
p@27 174 url={http://www.scopus.com/inward/record.url?eid=2-s2.0-84884332226&partnerID=40&md5=d028cb2aca5d2d6d8f25a8a8b555edbf},
p@27 175 document_type={Conference Paper},
p@27 176 source={Scopus},
p@27 177 }
p@27 178
p@27 179 @online{1_spotify_press_2014,
p@27 180 author={},
p@27 181 title={Spotify Acquires The Echo Nest},
p@27 182 url={https://press.spotify.com/us/2014/03/06/spotify-acquires-the-echo-nest/},
p@27 183 urldate={2015-8-27},
p@27 184 year={2014}
p@27 185 }
p@27 186 @article{smith2009social,
p@27 187 title={The social media revolution},
p@27 188 author={Smith, Tom},
p@27 189 journal={International journal of market research},
p@27 190 volume={51},
p@27 191 number={4},
p@27 192 pages={559--561},
p@27 193 year={2009}
p@27 194 }
p@27 195 @article{Putzke2014519,
p@27 196 title = "Cross-cultural gender differences in the adoption and usage of social media platforms – An exploratory study of Last.FM ",
p@27 197 journal = "Computer Networks ",
p@27 198 volume = "75, Part B",
p@27 199 number = "",
p@27 200 pages = "519 - 530",
p@27 201 year = "2014",
p@27 202 note = "Special Issue on Online Social NetworksThe Connectedness, Pervasiveness and Ubiquity of Online Social Networks ",
p@27 203 issn = "1389-1286",
p@27 204 doi = "http://dx.doi.org/10.1016/j.comnet.2014.08.027",
p@27 205 url = "http://www.sciencedirect.com/science/article/pii/S1389128614003302",
p@27 206 author = "Johannes Putzke and Kai Fischbach and Detlef Schoder and Peter A. Gloor",
p@27 207 keywords = "Adoption",
p@27 208 keywords = "Cross-cultural differences",
p@27 209 keywords = "Gender",
p@27 210 keywords = "Social media ",
p@27 211 abstract = "Abstract This paper examines cross-cultural gender differences in the adoption and usage of the social media platform Last.FM. From a large-scale empirical study of 3748 Last.FM users from Australia, Finland, Germany, and the United States of America, we find: (1) men listen to more pieces of music on social media platforms than do women; (2) women focus their listening on fewer musical genres and fewer tracks than do men; (3) women register on Last.FM later than do “early adopting” men (absolutely and in comparison to their friends), but at a younger age; (4) women maintain more virtual friendships on Last.FM than do men; and (5) women, when choosing music to listen to on social media platforms, are more likely than are men to choose tracks that correspond to mainstream tastes. "
p@27 212 }
p@27 213
p@27 214 @CONFERENCE{Park200811,
p@27 215 author={Park, Y.-J. and Tuzhilin, A.},
p@27 216 title={The long tail of recommender systems and how to leverage it},
p@27 217 journal={RecSys'08: Proceedings of the 2008 ACM Conference on Recommender Systems},
p@27 218 year={2008},
p@27 219 pages={11-18},
p@27 220 doi={10.1145/1454008.1454012},
p@27 221 note={},
p@27 222 url={http://www.scopus.com/inward/record.url?eid=2-s2.0-63449136183&partnerID=40&md5=648e50cac2d99764f891b5bc4b97bbfe},
p@27 223 document_type={Conference Paper},
p@27 224 source={Scopus},
p@27 225 }
p@27 226 @CONFERENCE{Dai20141760,
p@27 227 author={Dai, C. and Qian, F. and Jiang, W. and Wang, Z. and Wu, Z.},
p@27 228 title={A personalized recommendation system for netease dating site},
p@27 229 journal={Proceedings of the VLDB Endowment},
p@27 230 year={2014},
p@27 231 volume={7},
p@27 232 number={13},
p@27 233 pages={1760-1765},
p@27 234 note={},
p@27 235 url={http://www.scopus.com/inward/record.url?eid=2-s2.0-84905828317&partnerID=40&md5=90fbd8b20ad39757895bdee3ca58f459},
p@27 236 document_type={Article},
p@27 237 source={Scopus},
p@27 238 }
p@27 239
p@27 240 @CONFERENCE{Yin2012896,
p@27 241 author={Yin, H. and Cui, B. and Li, J. and Yao, J. and Chen, C.},
p@27 242 title={Challenging the long tail recommendation},
p@27 243 journal={Proceedings of the VLDB Endowment},
p@27 244 year={2012},
p@27 245 volume={5},
p@27 246 number={9},
p@27 247 pages={896-907},
p@27 248 note={},
p@27 249 url={http://www.scopus.com/inward/record.url?eid=2-s2.0-84863735354&partnerID=40&md5=2bd887772ba832fbbb4631afd25514d9},
p@27 250 document_type={Article},
p@27 251 source={Scopus},
p@27 252 }
p@27 253
p@27 254 @online{1_hypebot.com_2015,
p@27 255 author={Hypebot.com,},
p@27 256 title={Streaming Music Discovery: It's More Than Just Showing Album Credits - hypebot},
p@27 257 url={http://www.hypebot.com/hypebot/2015/07/streaming-music-discovery-its-more-than-just-showing-album-credits.html},
p@27 258 urldate={2015-8-26},
p@27 259 year={2015}
p@27 260 }
p@27 261
p@27 262 @online{ringen_2015,
p@27 263 author={Ringen, Jonathan},
p@27 264 title={Spotify, Apple Music, And The Streaming Wars: 5 Things We've Learned},
p@27 265 url={http://www.fastcompany.com/3048653/innovation-agents/listen-up},
p@27 266 urldate={2015-8-26},
p@27 267 journal={Fast Company},
p@27 268 year={2015}
p@27 269 }
p@27 270
p@27 271 @online{recsys2012,
p@27 272 author = {},
p@27 273 title = {Recommender Systems},
p@27 274 year = {2012},
p@27 275 url = {http://recommender-systems.org/},
p@27 276 note = {[Accessed: 26th August 2015]}
p@27 277 }
p@27 278
p@26 279 @article{Hejazi15,
p@26 280 author={Hejazi,S. A. and Stapleton,S. P.},
p@26 281 year={2015},
p@26 282 title={A self-organized network for load balancing using intelligent distributed antenna system},
p@26 283 journal={Canadian Journal of Electrical and Computer Engineering},
p@26 284 volume={38},
p@26 285 number={2},
p@26 286 pages={89-99},
p@26 287 url={www.scopus.com},
p@26 288 }
p@27 289
p@26 290
p@26 291 @book{larranaga2002estimation,
p@26 292 title={Estimation of distribution algorithms: A new tool for evolutionary computation},
p@26 293 author={Larranaga, Pedro and Lozano, Jose A},
p@26 294 volume={2},
p@26 295 year={2002},
p@26 296 publisher={Springer Science \& Business Media}
p@26 297 }
p@26 298
p@26 299 @incollection{melville2010recommender,
p@26 300 title={Recommender systems},
p@26 301 author={Melville, Prem and Sindhwani, Vikas},
p@26 302 booktitle={Encyclopedia of machine learning},
p@26 303 pages={829--838},
p@26 304 year={2010},
p@26 305 publisher={Springer}
p@26 306 }
p@26 307
p@20 308 @incollection{Lops2011,
p@20 309 year={2011},
p@20 310 isbn={978-0-387-85819-7},
p@20 311 booktitle={Recommender Systems Handbook},
p@20 312 editor={Ricci, Francesco and Rokach, Lior and Shapira, Bracha and Kantor, Paul B.},
p@20 313 doi={10.1007/978-0-387-85820-3_3},
p@20 314 title={Content-based Recommender Systems: State of the Art and Trends},
p@27 315 url={http://dx.doi.org/10.1007/978-0-387-85820-3_3},
p@20 316 publisher={Springer US},
p@20 317 author={Lops, Pasquale and de Gemmis, Marco and Semeraro, Giovanni},
p@20 318 pages={73-105},
p@20 319 language={English}
p@20 320 }
p@20 321
p@20 322 @ARTICLE{Burke2002331,
p@20 323 author={Burke, R.},
p@20 324 title={Hybrid recommender systems: Survey and experiments},
p@20 325 journal={User Modelling and User-Adapted Interaction},
p@20 326 year={2002},
p@20 327 volume={12},
p@20 328 number={4},
p@20 329 pages={331-370},
p@20 330 doi={10.1023/A:1021240730564},
p@27 331 note={},
p@27 332 url={http://www.scopus.com/inward/record.url?eid=2-s2.0-0036959356&partnerID=40&md5=28885a102109be826507abc2435117a7},
p@20 333 document_type={Article},
p@20 334 source={Scopus},
p@20 335 }
p@20 336
p@20 337 @ARTICLE{Yoshii2008435,
p@20 338 author={Yoshii, K. and Goto, M. and Komatani, K. and Ogata, T. and Okuno, H.G.},
p@20 339 title={An efficient hybrid music recommender system using an incrementally trainable probabilistic generative model},
p@20 340 journal={IEEE Transactions on Audio, Speech and Language Processing},
p@20 341 year={2008},
p@20 342 volume={16},
p@20 343 number={2},
p@20 344 pages={435-447},
p@20 345 doi={10.1109/TASL.2007.911503},
p@20 346 art_number={4432655},
p@27 347 note={},
p@27 348 url={http://www.scopus.com/inward/record.url?eid=2-s2.0-39649112098&partnerID=40&md5=6827f82844ae1da58a6fa95caf5092d9},
p@20 349 document_type={Article},
p@20 350 source={Scopus},
p@20 351 }
p@20 352
p@20 353
p@20 354 @article{JCC4:JCC4393,
p@20 355 author = {boyd, danah m. and Ellison, Nicole B.},
p@20 356 title = {Social Network Sites: Definition, History, and Scholarship},
p@20 357 journal = {Journal of Computer-Mediated Communication},
p@20 358 volume = {13},
p@20 359 number = {1},
p@20 360 publisher = {Blackwell Publishing Inc},
p@20 361 issn = {1083-6101},
p@20 362 url = {http://dx.doi.org/10.1111/j.1083-6101.2007.00393.x},
p@20 363 doi = {10.1111/j.1083-6101.2007.00393.x},
p@20 364 pages = {210--230},
p@20 365 year = {2007},
p@20 366 }
p@20 367
p@20 368 @ARTICLE{Castanedo2013,
p@20 369 author={Castanedo, F.},
p@20 370 title={A review of data fusion techniques},
p@20 371 journal={The Scientific World Journal},
p@20 372 year={2013},
p@20 373 volume={2013},
p@20 374 doi={10.1155/2013/704504},
p@20 375 art_number={704504},
p@27 376 note={},
p@27 377 url={http://www.scopus.com/inward/record.url?eid=2-s2.0-84888882639&partnerID=40&md5=827fabc750db24f662fdae1c798f2507},
p@20 378 document_type={Review},
p@20 379 source={Scopus},
p@20 380 }
p@20 381
p@20 382
p@20 383 @CONFERENCE{Lee20091096,
p@20 384 author={Lee, H. and Yan, L. and Pham, P. and Ng, A.Y.},
p@20 385 title={Unsupervised feature learning for audio classification using convolutional deep belief networks},
p@20 386 journal={Advances in Neural Information Processing Systems 22 - Proceedings of the 2009 Conference},
p@20 387 year={2009},
p@20 388 pages={1096-1104},
p@27 389 note={},
p@27 390 url={http://www.scopus.com/inward/record.url?eid=2-s2.0-84863380535&partnerID=40&md5=e872a6227c816850167f91bb2d41d8b7},
p@20 391 document_type={Conference Paper},
p@20 392 source={Scopus},
p@20 393 }
p@20 394
p@20 395
p@20 396 @TechReport {export:115396,
p@20 397 abstract = {<p>Recommender systems are now popular both commercially and in the research
p@20 398 community, where many approaches have been suggested for providing
p@20 399 recommendations. In many cases a system designer that wishes to employ a
p@20 400 recommendation system must choose between a set of candidate approaches. A first
p@20 401 step towards selecting an appropriate algorithm is to decide which properties of
p@20 402 the application to focus upon when making this choice. Indeed, recommendation
p@20 403 systems have a variety of properties that may affect user experience, such as
p@20 404 accuracy, robustness, scalability, and so forth. In this paper we discuss how to
p@20 405 compare recommenders based on a set of properties that are relevant for e
p@20 406 application. We focus on comparative studies, where a few algorithms are compared
p@20 407 using some evaluation metric, rather than absolute benchmarking of algorithms. We
p@20 408 describe experimental settings appropriate for making choices between algorithms.
p@20 409 We review three types of experiments, starting with an offline setting, where
p@20 410 recommendation approaches are compared without user interaction, then reviewing
p@20 411 user studies, where a small group of subjects experiment with the system and
p@20 412 report on the experience, and finally describe large scale online experiments,
p@20 413 where real user populations interact with the system. In each of these cases we
p@20 414 describe types of questions that can be answered, and suggest protocols for
p@20 415 experimentation. We also discuss how to draw trustworthy conclusions from e
p@20 416 conducted experiments. We then review a large set of properties, and explain how
p@20 417 to evaluate systems given relevant properties. We also survey a large set of
p@20 418 evaluation metrics in the context of the property that they evaluate.</p>},
p@20 419 author = {Guy Shani and Asela Gunawardana},
p@20 420 month = {November},
p@20 421 number = {MSR-TR-2009-159},
p@20 422 publisher = {Microsoft Research},
p@20 423 title = {Evaluating Recommender Systems},
p@20 424 url = {http://research.microsoft.com/apps/pubs/default.aspx?id=115396},
p@20 425 year = {2009},
p@20 426 }
p@20 427
p@20 428 @phdthesis {1242,
p@20 429 title = {Music Recommendation and Discovery in the Long Tail},
p@20 430 year = {2008},
p@20 431 school = {Universitat Pompeu Fabra},
p@20 432 address = {Barcelona},
p@20 433 abstract = {<p class="small">
p@20 434 Music consumption is biased towards a few popular artists. For instance, in 2007 only 1\% of
p@20 435 all digital tracks accounted for 80\% of all sales. Similarly, 1,000 albums accounted for 50\%
p@20 436 of all album sales, and 80\% of all albums sold were purchased less than 100 times. There is
p@20 437 a need to assist people to filter, discover, personalise and recommend from the huge amount
p@20 438 of music content available along the Long Tail.
p@20 439 </p>
p@20 440 <p class="small">
p@20 441 Current music recommendation algorithms try to
p@20 442 accurately predict what people demand to listen to. However, quite
p@20 443 often these algorithms tend to recommend popular -or well-known to the
p@20 444 user- music, decreasing the effectiveness of the recommendations. These
p@20 445 approaches focus on improving the accuracy of the recommendations. That
p@20 446 is, try to make
p@20 447 accurate predictions about what a user could listen to, or buy next,
p@20 448 independently of how
p@20 449 useful to the user could be the provided recommendations.
p@20 450 </p>
p@20 451 <p class="small">
p@20 452 In this Thesis we stress the importance of the user{\textquoteright}s
p@20 453 perceived quality of the recommendations. We model the Long Tail curve
p@20 454 of artist popularity to predict -potentially-
p@20 455 interesting and unknown music, hidden in the tail of the popularity
p@20 456 curve. Effective recommendation systems should promote novel and
p@20 457 relevant material (non-obvious recommendations), taken primarily from
p@20 458 the tail of a popularity distribution.
p@20 459 </p>
p@20 460 <p class="small">
p@20 461 The main contributions of this Thesis are: <em>(i)</em> a novel network-based approach for
p@20 462 recommender systems, based on the analysis of the item (or user) similarity graph, and the
p@20 463 popularity of the items, <em>(ii)</em> a user-centric evaluation that measures the user{\textquoteright}s relevance
p@20 464 and novelty of the recommendations, and <em>(iii)</em> two prototype systems that implement the
p@20 465 ideas derived from the theoretical work. Our findings have significant implications for
p@20 466 recommender systems that assist users to explore the Long Tail, digging for content they
p@20 467 might like.
p@20 468 </p>
p@20 469 },
p@27 470 url = {http://mtg.upf.edu/static/media/PhD_ocelma.pdf},
p@27 471 author = {Celma, \`{O}.}
p@20 472 }
p@21 473
p@21 474 @inproceedings{pachet2001musical,
p@21 475 title={Musical data mining for electronic music distribution},
p@21 476 author={Pachet, Fran{\c{c}}ois and Westermann, Gert and Laigre, Damien},
p@21 477 booktitle={Web Delivering of Music, 2001. Proceedings. First International Conference on},
p@21 478 pages={101--106},
p@21 479 year={2001},
p@21 480 organization={IEEE}
p@21 481 }
p@21 482
p@21 483 @ARTICLE{Tzanetakis2002293,
p@21 484 author={Tzanetakis, G. and Cook, P.},
p@21 485 title={Musical genre classification of audio signals},
p@21 486 journal={IEEE Transactions on Speech and Audio Processing},
p@21 487 year={2002},
p@21 488 volume={10},
p@21 489 number={5},
p@21 490 pages={293-302},
p@21 491 doi={10.1109/TSA.2002.800560},
p@27 492 note={},
p@27 493 url={http://www.scopus.com/inward/record.url?eid=2-s2.0-0036648502&partnerID=40&md5=72d2fee186b42c9998f13415cbb79eea},
p@21 494 document_type={Article},
p@21 495 source={Scopus},
p@21 496 }
p@21 497
p@21 498 @CONFERENCE{Sturm20127,
p@21 499 author={Sturm, B.L.},
p@21 500 title={An analysis of the GTZAN music genre dataset},
p@21 501 journal={MIRUM 2012 - Proceedings of the 2nd International ACM Workshop on Music Information Retrieval with User-Centered and Multimodal Strategies, Co-located with ACM Multimedia 2012},
p@21 502 year={2012},
p@21 503 pages={7-12},
p@21 504 doi={10.1145/2390848.2390851},
p@27 505 note={},
p@27 506 url={http://www.scopus.com/inward/record.url?eid=2-s2.0-84870497334&partnerID=40&md5=40a48c1c9d787308dd315694b54b64ec},
p@21 507 document_type={Conference Paper},
p@21 508 source={Scopus},
p@21 509 }
p@21 510
p@21 511 @unpublished{Bengio-et-al-2015-Book,
p@21 512 title={Deep Learning},
p@21 513 author={Yoshua Bengio and Ian J. Goodfellow and Aaron Courville},
p@21 514 note={Book in preparation for MIT Press},
p@21 515 url={http://www.iro.umontreal.ca/~bengioy/dlbook},
p@21 516 year={2015}
p@21 517 }
p@21 518
p@21 519 @CONFERENCE{Sigtia20146959,
p@21 520 author={Sigtia, S. and Dixon, S.},
p@21 521 title={Improved music feature learning with deep neural networks},
p@21 522 journal={ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings},
p@21 523 year={2014},
p@21 524 pages={6959-6963},
p@21 525 doi={10.1109/ICASSP.2014.6854949},
p@21 526 art_number={6854949},
p@27 527 note={},
p@27 528 url={http://www.scopus.com/inward/record.url?eid=2-s2.0-84905259152&partnerID=40&md5=3441dfa8c7998a8eb39f668d43efb8a1},
p@21 529 document_type={Conference Paper},
p@21 530 source={Scopus},
p@21 531 }
p@21 532
p@21 533 @incollection{NIPS2013_5004,
p@21 534 title = {Deep content-based music recommendation},
p@21 535 author = {van den Oord, Aaron and Dieleman, Sander and Schrauwen, Benjamin},
p@21 536 booktitle = {Advances in Neural Information Processing Systems 26},
p@21 537 editor = {C.J.C. Burges and L. Bottou and M. Welling and Z. Ghahramani and K.Q. Weinberger},
p@21 538 pages = {2643--2651},
p@21 539 year = {2013},
p@21 540 publisher = {Curran Associates, Inc.},
p@21 541 url = {http://papers.nips.cc/paper/5004-deep-content-based-music-recommendation.pdf}
p@21 542 }
p@21 543
p@21 544 @incollection{pelikan2015estimation,
p@21 545 title={Estimation of Distribution Algorithms},
p@21 546 author={Pelikan, Martin and Hauschild, Mark W and Lobo, Fernando G},
p@21 547 booktitle={Springer Handbook of Computational Intelligence},
p@21 548 pages={899--928},
p@21 549 year={2015},
p@21 550 publisher={Springer}
p@21 551 }
p@21 552
p@21 553 @article{Santana:Bielza:Larrañaga:Lozano:Echegoyen:Mendiburu:Armañanzas:Shakya:2009:JSSOBK:v35i07,
p@21 554 author = "Roberto Santana and Concha Bielza and Pedro Larrañaga and Jose A. Lozano and Carlos Echegoyen and Alexander Mendiburu and Rubén Armañanzas and Siddartha Shakya",
p@21 555 title = "Mateda-2.0: A MATLAB Package for the Implementation and Analysis of Estimation of Distribution Algorithms",
p@21 556 journal = "Journal of Statistical Software",
p@21 557 volume = "35",
p@21 558 number = "7",
p@21 559 pages = "1--30",
p@21 560 day = "26",
p@21 561 month = "7",
p@21 562 year = "2010",
p@21 563 CODEN = "JSSOBK",
p@21 564 ISSN = "1548-7660",
p@21 565 bibdate = "2009-12-17",
p@21 566 URL = "http://www.jstatsoft.org/v35/i07",
p@21 567 accepted = "2009-12-17",
p@21 568 acknowledgement = "",
p@21 569 keywords = "",
p@21 570 submitted = "2009-04-15",
p@21 571 }
p@21 572
p@21 573 @ARTICLE{Liang2014781,
p@21 574 author={Liang, T. and Liang, Y. and Fan, J. and Zhao, J.},
p@21 575 title={A hybrid recommendation model based on estimation of distribution algorithms},
p@21 576 journal={Journal of Computational Information Systems},
p@21 577 year={2014},
p@21 578 volume={10},
p@21 579 number={2},
p@21 580 pages={781-788},
p@21 581 doi={10.12733/jcis9623},
p@27 582 note={},
p@27 583 url={http://www.scopus.com/inward/record.url?eid=2-s2.0-84892865461&partnerID=40&md5=a2927d36b493e8ef4d1cdab3055fa68b},
p@21 584 document_type={Article},
p@21 585 source={Scopus},
p@21 586 }
p@21 587
p@21 588 @INPROCEEDINGS{Bertin-Mahieux2011,
p@21 589 author = {Thierry Bertin-Mahieux and Daniel P.W. Ellis and Brian Whitman and Paul Lamere},
p@21 590 title = {The Million Song Dataset},
p@21 591 booktitle = {{Proceedings of the 12th International Conference on Music Information
p@21 592 Retrieval ({ISMIR} 2011)}},
p@21 593 year = {2011},
p@21 594 owner = {thierry},
p@21 595 timestamp = {2010.03.07}
p@21 596 }
p@21 597
p@21 598 @inproceedings{zhang2014deep,
p@21 599 title={A deep representation for invariance and music classification},
p@21 600 author={Zhang, Chiyuan and Evangelopoulos, Georgios and Voinea, Stephen and Rosasco, Lorenzo and Poggio, Tomaso},
p@21 601 booktitle={Acoustics, Speech and Signal Processing (ICASSP), 2014 IEEE International Conference on},
p@21 602 pages={6984--6988},
p@21 603 year={2014},
p@21 604 organization={IEEE}
p@21 605 }