3.4.3 Heterogeneous Preference Embedding
HPE [5] ⌥⌦i⇧Â\¯<F ·¯ ü,HPE ø/Â˙ºÂb✏ query intention Ñ®¶OLZ∫ñÅÑ˚ŸÓ⇡ @Â÷Ñ≤Ôh:’x“fl⌦i
⇧Â\¯‘ ZÜ õÙ’⇥HPE ;Å/9⁄(⇧ÑO}Ü2Lx“ @Â
↵M’ Asynchronous Stochastic Gradient Descent 3.9✏2L s ⇥
3.5 ® ® ®¶ ¶ ¶˚ ˚ ˚q q q
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Scoreu,q,i= p⇥ (ru· rTi ) + (1 p)⇥ (rq · rTi ) (3.10) v- q, i2 |I| u2 |U| p 2 [0, 1] Kå⌘⌘ç9⁄d⌃x ÷˙Mk ñ íè⌃x ÿÑLÚ ⇣↵∫ Ñ®¶⌫h⇥
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
, , , € € € ‡ ‡ ‡ Ê
Ê ÊW W WP P Pú ú ú⌥ ⌥ ⌥ ÷ ÷ ÷
(d‡¿ ⌘⌘⌥⇤ ÷⌥⌃ê⌘⌘Ñ∂Àåπ’@ó0Ñh˛Pú⇥(4.1✏¿
-⌘⌘⇤X ↵⌘⌘@(Ñ«ô∆ v«ôy' &U:v¯‹Ñq x⁄⇥ (4.2✏¿ ⌘⌘⇤™ &–õ ÊWÑ!ã@(Ñ√x ⌘⌘Ñ‘
⇤π’⇥(4.3✏¿ ⌘⌘G⇤À9⌘⌘@(óU0⌥ ⇡⇥( 4.4✏¿- ⌘⌘
⇤‘⇤ ⌃ê⌘⌘π’⌥v÷π’Ñh˛⇥ å (4.5- ⌘⌘⇤✏NÊãÑU :Ü¿fld˚qÑh˛⇥
4.1 « « «ô ô ô∆ ∆ ∆
(d«÷á-⌘⌘( KKBOX ºÛ⇥2AsÑF} ⌅\∫«ô∆ÜÊ W⇥ KKBOX 6∆Ü@ (6(Í∂Û⇥2As-@Z˙ÑF}L∫ œ
F«ô˝s00⇠⌅Ü ↵(6(¿ºBìfi}ÜÍ ñL ûÍ.›n{
e F}Bì 0@⇥,÷áx˙Ü∞ Ñ A ⇥4\∫Ó⇡«ô∆ U⌃
Ü2018 t 5 Û8 ºA ⇥4Ñ@ F} ⌅ &⌥ 5 Û 7 Ñ«ôvZ◆Ù
«ô∆ Â↵!1◆Ù∆A 8 Ñ«ôvZ,f«ô∆ Â↵!1,f
∆A Ê πb ⌘⌘_x˙Ü⌘⌘(ÜP Pë⌦ÑB ⇥4\∫Ü ê«ô∆ #_U⌃Ü2018 t 5 Û7 Ñ«ôvZ◆Ù∆ Â↵!1◆Ù
∆B ⇥
h4.1: A B⇥4«ô∆x⁄q
‧
|U| 1580 1548 249640
|I| 22170 12276 558601
|S| 63938 31848 22258342
t(·) 40.47 20.57 89.16
k(·) 288.40 111,48 629.40
è¶ 0.182% 0.168% 0.016%
v- |S| ∫@ (6 - LÚ Õ⌥KF}D t(·) ∫(6sGF}ÑLÚ
-öh:’Ñ≠¶ Dimensions ∫ 100 œ↵¿fivZwfiÑ!x Walk Times ∫ 20 ! œ!JpÑ›‚ Walk Step ∫ 40 e föÑ ñó'✏ Window Size ∫ 3 œ!†Ω#Ñxœ Negative Samples
∫5 !⇥
‧
• HPE⇢⇢⇢Heterogeneous Preference Embedding ÔÂ⌥(⇧ÑpÍ«⌦P w Üx“ &9⁄x“0Ñ(⇧ÑO}2L¯‹Ñ®¶⇥ ⌘⌘-öHPE h
• Superhighway+DeepWalk⇢⇢⇢(Ö#P \˙ - ⌘⌘-öi↵(6Kì q ÑLÚ‘ã ↵ <∫0.9 ˙ÀÖ#PÑ⌦Õ ∫1.0 ⇥ (◆Ù
‧
‧
‧
Method P@10 P@20 P@30 P@10 Improvement Deepwalk 0.0402 0.0337 0.0302
-Superhighway+
Deepwalk 0.0552 0.0430 0.0372 37.3%
LINE 0.0528 0.0438 0.0387 -Superhighway+
LINE 0.0522 0.0425 0.0356 -1.1%
HPE 0.0407 0.0333 0.0298 -Superhighway+
HPE 0.0775 0.0614 0.0525 90.4%
h4.3: ñ∫áh˛
‧
Method R@10 R@20 R@30 R@10 Improvement Deepwalk 0.0528 0.0540 0.0576
-Superhighway+
Deepwalk 0.0670 0.0646 0.0681 26.9%
LINE 0.0660 0.0693 0.0782 -Superhighway+
LINE 0.0628 0.0675 0.0702 -5.1%
HPE 0.0518 0.0543 0.0595 -Superhighway+
HPE 0.0889 0.0915 0.0978 71.6%
h4.4: Ïfiáh˛
Method MAP@10 MAP@20 MAP@30 MAP@10 Improvement Deepwalk 0.0295 0.0254 0.0246
-Superhighway+
Deepwalk 0.0346 0.0276 0.0253 17.2%
LINE 0.0358 0.0304 0.0291 -Superhighway+
LINE 0.0367 0.0292 0.0266 2.5%
HPE 0.0292 0.0262 0.0252
-Superhighway+
HPE 0.0509 0.0410 0.0379 74.3%
h4.5: sGñ∫áG<h˛
1h4.5-⌘⌘ÔÂ↵0 k = 10 ÑB⇡ ⌘⌘Ñ˙ π’((⌦ .
≤Ôh:’⌦ h˛˝‘ü,Ñπ’Å}⇥ HPE Í(ñ∫á Ïfiá⌦ o WÑ–G (sGñ∫áG<h˛⌦_à} h:HPE (⌘⌘˙ π’Ñk©
↵ Í˝ Ù& (⇧Ñ B~0 £ÑLÚ ⇡õ ¯‹ ÑLÚ_'
‧
Method Novelty@10 Novelty@20 Novelty@30 Deepwalk 0.666 0.671 0.670 Superhighway+
Deepwalk 0.710 0.721 0.726
LINE 0.587 0.617 0.631
Superhighway+
LINE 0.690 0.700 0.710
HPE 0.746 0.750 0.750
Superhighway+
HPE 0.613 0.622 0.632
h4.6: Top-N ∞N¶h˛
¯ Ñ®¶LÆÜ™ ∞N↵¶⇤ÿÑLÆ_„hW(& (⇧ú}ÑP6K
↵˝ WÙ⇢# ÑLÚ _˝ ö↵¶⌦k©(⇧¢"v÷*ÂÑLÚ⇥Ç
⌦h@U:Ñ (Deepwalk fl LINE ⌦ ⌘⌘Ñπ’(∞N¶⌦˝›Nü,Ñπ
’ F(HPE ⌦ ⌘⌘Ñπ’(∞N¶⌦Õ h˛ s⇥⌘⌘ç∫ü‡/‡∫∞
N¶(–.↵¶⌦/⌥ñ∫á Ïfiá sGñ∫áG< .⇡ñ ↵¯ís aÑ‹o‹¬ ãÇ⌘⌘ÛìsGñ∫áG<˝–G F–GÑq 1h:LÆ g(⇧í }NÑL !¯‹ ‘á1⇤↵M _1 ⌘(⇧¢" ”UÑ _⇤⇥HPE Ñh˛1h:W⇡ fi LINE ÷6(ñ∫áIU0⌦H˛↵M®
‚ F(∞N¶⌦{›Nü,Ñπ’ ‰⌘⌘Z Ñ/Deepwalk ≈( ⇧ U0⌦h˛›Nü,Ñπ’ (∞N¶⌦h˛_⇤s⇥
ú Â⌦U0⇡ñ ⌘⌘ç∫⌘⌘Ñπ’˝ ✏NÖ#PÑ˙äÜk©⌅^
≤Ôh:’x“π’⌥≤Ôx“ÑÙ} & H˜÷(⇧O}Âó0Ù ¡Í
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Ñ≤Ôh:’ÜZ®¶⇥
4.5 Ê Ê Êã ã ã⌃ ⌃ ⌃ê ê ê
4.5.1 Ê Ê Êã ã ã⌃ ⌃ ⌃ê ê ê -  ® ® ®¶ ¶ ¶˚ ˚ ˚q q q∫ ∫ ∫ã ã ã
(d✏¿- ⌘⌘✏NÊõÑãPÜU:⌘⌘Ñ®¶˚q(ÊõÑ®¶OL⌦/
&˝ & ⌘⌘Ñ⇣  ⌦π’å†e⌘⌘Ñπ’(x“⌦ÑPú⇥
∫ÜÅ⌃ê⌘⌘®¶ÑPú/&& ⌘⌘Ñ ⌘⌘ºÓ⇡«ô∆-⌘x Ü (⇧ &)(÷ÑwÚF} ⌅Ü˙À®¶Ñ⇧Æ⇥h4.7⌫˙⌘⌘
( π’Ñ!ã@®¶˙ÜÑPú⇥9⁄⇡ (⇧÷⌘wÚ ⌅Ü®∑
⇡ (⇧àúaF}80t„ÑAL ˛⇥Ú ⌘⌘⇣ ®¶˚q_…r
⇤®¶÷ 80 t„ÑAL / ˛LÚ⇥ (ü,Ñ𒮶↵ LÆ·b_˜
‹Üœ/For Your Entertainment Oh My My ⇡^˛„LÚ ⇢Û/ùªWÑL ÚPrincess and the Pea II ⌘⌘Ñπ’G~N˝⇤®fi80 t„ÑALLK
˛⇥⇠ ⇡_/& ⌘⌘ Ñ®¶⇥1dÊãÔÂÂS⌘⌘Ñπ’ÔÂ˙º T N˛Â w˚ÂXÜóLÚÑyµç (≤Ôh:’·b
4.5.2 Ê Ê Êã ã ã⌃ ⌃ ⌃ê ê ê -  Â≤ ≤ ≤Ô Ô Ôh h h: : :’ ’ ’x x x“ “ “∫ ∫ ∫ã ã ã
(⌘⌘Ñ®¶˚qK- ≤Ôh:’x“Ñ≈¡⇤àÙ•ÑqˇLÚÑ®¶H
ú º(6Ñ≤Ôh:’_˝9⁄ Å2L¯‹Ñ⌃ê≈É⇥✏Nh:’x
“ ⌘⌘ ⇢
• ˝ ✏NÜê«⌦ÑÂX eÜ„zÓ⇡«ôÑ è'OL
• ºLÚÜ™ ®<^<ÑLÚ…rÅ䯗ä}⇥_1/ LÚÑ≤Ô h:✏ (6ìT N˛Ñyµ(·b
Â↵ ⌘⌘(Selena Gomez Ñ Fetish ⇡ñLZ∫ÂbÑLÚ › π✏↵
Ñ≤Ôh:’2L⌃ê⇥
h4.8 ⌥h 4.9∫⌘⌘)( Fetish ⇡ñLÑ≤Ôh:’( Ñπ’ï↵2L
⇠&¯<¶Ñ⌧↵@~˙ÜÑ 10 ñL ÔÂ↵0 (ü,Ñπ’K- dÜ~
0ÜœDemi Lovato Ariana Grande ⇡^⌥ Selena Gomez ¯<ÑsLK®<ÚK
<Nј‹Ü-áí≈ ”áAL⇢Û/R LÚI®<ÂpÑLÚ h:
‧
(⇧F}⇠⌅ Careless Whisper George Michael Sweet Child O’ Mine Guns N’ Roses Wake Me Up Before You Go-Go George Michael
Method LÚ 1 LK
Deepwalk Can’t Fight This Feeling Essential - 80’s Livin’ On A Prayer Bon Jovi Rock You Like A Hurricane Scorpions
Bodies Drowning Pool For Your Entertainment Adam Lambert LINE Roll With It Backstreet Boys
Oh My My JTR
Nobody But You Backstreet Boys I Believe Elvis Presley Every Morning Sugar Ray
HPE Bodies Drowning Pool
I’m With You Bon Jovi Never Say Goodbye Bon Jovi Pretty Fly for a White Guy The Offspring
Princess and the Pea Sesame Street
Superhighway+ Carrie Europe
Deepwalk Glory Of Love Peter Cetera Total Eclipse Of The Heart Bonnie Tyler
Against All Odds Phil Collins Cheri Cheri Lady Modern Talking Superhighway+ Last Christmas George Michael
LINE Carrie Europe
Desperado Eagles The Final Countdown Europe
Fastlove George Michael
Superhighway+ Carrie Europe
HPE Without You Air Supply
Against All Odds Phil Collins Always Bon Jovi Making Love Out Of Nothing At All Air Supply
h4.7: ®¶˚qKÊã
‧
query Fetish Selena Gomez
Method LÚ 1 LK
Deepwalk My Everything Ariana Grande Cry Baby Demi Lovato Dark Side Bishop Briggs
q ✏◆}⌧Ë Xun
Eye of the Needle Sia
Reaper Sia
Unstoppable Sia
NOMAD Walk Off The Earth Rainbow Kesha
2! 3! BANGTAN BOYS LINE Issues Julia Michaels
Chained To The Rhythm Katy Perry Cry Baby Demi Lovato
Rainbow Kesha Neon Lights Demi Lovato Same Old Love Selena Gomez
Ikuyo Kyle
Who Says Selena Gomez Bed Nicki Minaj Impossible Shontelle
HPE Cry Baby Demi Lovato
Rainbow Kesha Dark Side Bishop Briggs Eye of the Needle Sia
q ✏◆}⌧Ë Xun
NOMAD Walk Off The Earth Mr Lonely (feat. Fat Lip) Portugal. The Man
My Everything Ariana Grande 2! 3! BANGTAN BOYS Stone Cold Demi Lovato h4.8: ≤Ôh:’x“Êã K
‧
query Fetish Selena Gomez
Method LÚ 1 LK
Superhighway+ Bad Liar Selena Gomez Deepwalk The Way I Are (Dance With Somebody) Bebe Rexha
Sober Selena Gomez Crying in the Club Camila Cabello
Good For You Selena Gomez
Woman Kesha
Call It What You Want Taylor Swift Kill Em With Kindness Selena Gomez The Heart Wants What It Wants Selena Gomez Remember I Told You Nick Jonas Superhighway+ Bad Liar Selena Gomez LINE The Way I Are (Dance With Somebody) Bebe Rexha
Sober Selena Gomez Call It What You Want Taylor Swift
Slow Down Selena Gomez Down Fifth Harmony The Heart Wants What It Wants Selena Gomez Crying in the Club Camila Cabello
Woman Kesha
Gorgeous Taylor Swift Superhighway+ Crying in the Club Camila Cabello
HPE Bad Liar Selena Gomez
The Way I Are (Dance With Somebody) Bebe Rexha Back to You Louis Tomlinson
Friends Justin Bieber, BloodPopR I Got You Bebe Rexha Swish Swish Katy Perry ...Ready For It? Taylor Swift
Down Fifth Harmony Gorgeous Taylor Swift h4.9: ≤Ôh:’x“Êã Kå
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
(x“h:’ÑB⇡Ñ/⇤ ö↵¶Ñ◊0«ô è'qˇ ⇤ OÓ ⌘⌘
Ñπ’@~fiÜÑG~N˝/®<¯—ÑLÚ ⇢Û˝/s'LKE⇢ ⇡‘⇤
& ⌘⌘⇣ ÑPú _IÊÜw˚Üê«ôÑÂX0Ó⇡«ô/ ©º(Ó⇡
«ô⌦Ñx“⇥
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
, , , î î î ‡ ‡ ‡ P
P P÷ ÷ ÷
(,÷á- º ↵«⌦⇢*⇣üÑÛ⇥2A ŸÑ®¶˚q ⌘⌘–˙Ü
↵i(Ñ˙ Ve ✏NÜê⇥4-⇤∫PÃÑ(⇧F}⇠⌅⌥Ó⇡⇥4Ñ«
⌦ZP K(ÜÖ#P \ÑVe 9⁄vq´ÑLÚ∆ ⌥Üê⌥Ó⇡Ñ
(⇧¿fiZ#P ç⌥ \ÂpÍ'≤Ôh:’Zx“ óÜê«⌦ÑÂX
˝ ⌃0w˚ÛÓ⇡«⌦ 2 9ÑÓ⇡«⌦Ñ≤Ôh:’ 2 å⇣˙ºÂb Ñ®¶⇥⌥ü Ñπ’Z‘⇤ ⌘⌘Ñπ’(sGñ∫áG<IU0x⁄⌦Ü /Ñh˛ (Û⇥2A Ÿ-_àËÕÑ∞N¶h˛⌦_÷óÜà}Ñsa⇥
ÂÊõK(⌦Ü↵ ⌘⌘Ñπ’˝ («ô ≥Ñ≈¡↵ …1w˚M «⌦Ñ ÂX2Ü h˛T0–G _˝ 9Ñ˛Ê®¶˚q8ãÑ«ô è'ÑOL⇥
(*ÜU ⌦ ⌘⌘⇤⇤n( Ñ≤Ôh:’ÜZx“ _ÔÂ⌥≤Ô h:’x“0Ñh:✏v⇣LÚ (6Ñyµ<ç(v÷!ãZ◆Ù⇥Ê
ºLÚÑ⌅⇧v÷l'Ñ«⌦ÇLK \Ú⇧ L^ Ú®IIvZyµÜ w◆Ù_/ÔÂ⌫fÑπ⌘K ⇥çÜ ÙÔÂP LÚÛ⌦I⇢í‘«⌦ P
—tܱÄÑÒ¶x“∂À 1˝û-˜÷0 (Ñ«⌦⇥
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
√
√ √⇤ ⇤ ⇤á á á{ { {
[1] G. Adomavicius and A. Tuzhilin. Towards the Next Generation of Recommender Systems: A Survey of the State-of-the-Art and Possible Extensions. IEEE Transac-tions on Knowledge and Data Engineering, 17(6):734–749, June 2005.
[2] Y. Bengio, A. Courville, and P. Vincent. Representation learning: A review and new perspectives. IEEE Trans. Pattern Anal. Mach. Intell., 35(8):1798–1828, Aug. 2013.
[3] B. Bocsi, L. Csat´o, and J. Peters. Alignment-based transfer learning for robot mod-els. pages 1–7, 08 2013.
[4] R. Burke. Hybrid recommender systems: Survey and experiments. User Modeling and User-Adapted Interaction, 12(4):331–370, Nov. 2002.
[5] C.-M. Chen, M.-F. Tsai, Y.-C. Lin, and Y.-H. Yang. Query-based music recom-mendations via preference embedding. In Proceedings of the 10th ACM Conference on Recommender Systems, RecSys ’16, pages 79–82, New York, NY, USA, 2016.
ACM.
[6] C.-M. Chen, Y.-H. Yang, Y. Chen, and M.-F. Tsai. Vertex-context sampling for weighted network embedding. arXiv preprint arXiv:1711.00227, 2017.
[7] P. Cremonesi, A. Tripodi, and R. Turrin. Cross-domain recommender systems. In Proceedings of the 2011 IEEE 11th International Conference on Data Mining Work-shops, ICDMW ’11, pages 496–503, Washington, DC, USA, 2011. IEEE Computer Society.
[8] W. Dai, Q. Yang, G.-R. Xue, and Y. Yu. Self-taught clustering. In Proceedings of the 25th International Conference on Machine Learning, ICML ’08, pages 200–207, New York, NY, USA, 2008. ACM.
[9] Y. Ganin and V. Lempitsky. Unsupervised domain adaptation by backpropagation.
In Proceedings of the 32Nd International Conference on International Conference on Machine Learning - Volume 37, ICML’15, pages 1180–1189. JMLR.org, 2015.
‧
recom-mendation without sharing user-relevant data. In The World Wide Web Conference, WWW ’19, pages 491–502, New York, NY, USA, 2019. ACM.[11] A. Grover and J. Leskovec. Node2vec: Scalable feature learning for networks. In Proceedings of the 22Nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD ’16, pages 855–864, New York, NY, USA, 2016.
ACM.
[12] Y. Hu, Y. Koren, and C. Volinsky. Collaborative filtering for implicit feedback datasets. In Proceedings of the 2008 Eighth IEEE International Conference on Data Mining, ICDM ’08, pages 263–272, Washington, DC, USA, 2008. IEEE Computer Society.
[13] Y. Koren, R. Bell, and C. Volinsky. Matrix factorization techniques for recommender systems. Computer, 42(8):30–37, Aug. 2009.
[14] K. Lai, T. Wang, H. Chi, Y. Chen, M. Tsai, and C. Wang. Superhighway: Bypass data sparsity in cross-domain CF. CoRR, abs/1808.09784, 2018.
[15] B. Li, Q. Yang, and X. Xue. Can movies and books collaborate?: Cross-domain collaborative filtering for sparsity reduction. In Proceedings of the 21st Interna-tional Jont Conference on Artifical Intelligence, IJCAI’09, pages 2052–2057, San Francisco, CA, USA, 2009. Morgan Kaufmann Publishers Inc.
[16] J. McAuley, C. Targett, Q. Shi, and A. van den Hengel. Image-based recommen-dations on styles and substitutes. In Proceedings of the 38th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR
’15, pages 43–52, New York, NY, USA, 2015. ACM.
[17] T. Mikolov, I. Sutskever, K. Chen, G. Corrado, and J. Dean. Distributed represen-tations of words and phrases and their compositionality. In Proceedings of the 26th International Conference on Neural Information Processing Systems - Volume 2, NIPS’13, pages 3111–3119, USA, 2013. Curran Associates Inc.
[18] S. J. Pan and Q. Yang. A survey on transfer learning. IEEE Trans. on Knowl. and Data Eng., 22(10):1345–1359, Oct. 2010.
[19] M. J. Pazzani and D. Billsus. The adaptive web. chapter Content-based Recommen-dation Systems, pages 325–341. Springer-Verlag, Berlin, Heidelberg, 2007.
[20] B. Perozzi, R. Al-Rfou, and S. Skiena. Deepwalk: Online learning of social repre-sentations. In Proceedings of the 20th ACM SIGKDD International Conference on
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Knowledge Discovery and Data Mining, KDD ’14, pages 701–710, New York, NY, USA, 2014. ACM.
[21] R. Raina, A. Battle, H. Lee, B. Packer, and A. Y. Ng. Self-taught learning: Trans-fer learning from unlabeled data. In Proceedings of the 24th International ConTrans-fer- Confer-ence on Machine Learning, ICML ’07, pages 759–766, New York, NY, USA, 2007.
ACM.
[22] S. Rendle. Factorization machines. In Proceedings of the 2010 IEEE International Conference on Data Mining, ICDM ’10, pages 995–1000, Washington, DC, USA, 2010. IEEE Computer Society.
[23] S. Rendle, C. Freudenthaler, Z. Gantner, and L. Schmidt-Thieme. Bpr: Bayesian personalized ranking from implicit feedback. In Proceedings of the Twenty-Fifth Conference on Uncertainty in Artificial Intelligence, UAI ’09, pages 452–461, Ar-lington, Virginia, United States, 2009. AUAI Press.
[24] P. Resnick, N. Iacovou, M. Suchak, P. Bergstrom, and J. Riedl. Grouplens: An open architecture for collaborative filtering of netnews. pages 175–186. ACM Press, 1994.
[25] J. Tang, M. Qu, M. Wang, M. Zhang, J. Yan, and Q. Mei. Line: Large-scale infor-mation network embedding. In Proceedings of the 24th International Conference on World Wide Web, WWW ’15, pages 1067–1077, Republic and Canton of Geneva, Switzerland, 2015. International World Wide Web Conferences Steering Committee.
[26] Y. Zhang, Q. Ai, X. Chen, and W. B. Croft. Joint representation learning for top-n recommetop-ndatiotop-n with heterogetop-neous itop-nformatiotop-n sources. Itop-n Proceeditop-ngs of the 2017 ACM on Conference on Information and Knowledge Management, CIKM ’17, pages 1449–1458, New York, NY, USA, 2017. ACM.