Know your neighbours: Machine Learning on Graphs
|
|
- Aubrie Chandler
- 5 years ago
- Views:
Transcription
1 Know your neighbours: Machine Learning on Graphs Andrew Docherty Senior Research Engineer
2 2 Graphs are Everywhere Online Social Networks Transportation Networks Power Networks
3 Networks are Everywhere Topic models Financial transactions Terrorist Networks 3 Valdis Krebs,
4 4 What is a graph/network? node / vertex graph / network
5 5 What is a graph/network? node / vertex edge / link graph / network
6 6 What is a graph/network? Node types Heterogeneous graph / network Edge types
7 7 <latexit sha1_base64="p2ztnnfntkipyhj1fz96crcv/ug=">aaacfxicbzdlssnafiyn9vbrrerszwarxwhirpacqsgnywrgfprstqbtduhkemzoxbl6fg58ftcuvnwk7nwbp20w2npg4op/z+hm+ynyca2o823l5uyxfpfyy4wv1bx1jelm1p2oekwzrymrqvpanbncmg84cfalfsnhifg16f+n/oo9u5ph8hygmwuepct5h1mcrmowj0ht4kvsa3ua1e9bipo4+xfytu1d3cjdnmhsd4fnysmxnxhhwxazkkgsks3il9+oabiycvqqreuue0mjjqo4fwxy8bpnykl7zkvdocqh0410fnyq7xmljturmk8chqu/j1isaj0ia9mzeujpaw8k/ufve+icnviu4wsypjnfnurgipaoi9zmileqawoekm7+immpkelbjfkwibjtj8+cd2yf2+7nsalcztliox20iw6qi05rgv2jcviqry/ogb2in+vjerhery9ja87kzrbrn7i+fwbnx5x7</latexit> <latexit sha1_base64="d8xmnb8j/4th+o9jv8ogl9cpwge=">aaacbxicbvdjsgnbeo2jw4xb1kmijuh0nmyi4ajcwivhcmyemih2dcpjk56f7hoxdhpy4q948adi1x/w5t/ywq6a+kdg8v4vvfx8waqnjvnt5ebmfxax8sufldw19y3i5tatjhlfocojgam6zzrieuivbuqoxwpy4euo+f3lov+7b6vffn7giizmwlqh6ajo0eit4i5rab2ghsidpl56hykag3nq2zb1sqxvldm2mwkdje6elmgelvbxy2thpakgrc6z1g3xibgzmowcs8gkxqihzrzputawngqb6gy6eioj+0zp006ktivir+rvizqfwg8c33qgdht62huk/3mnbdunzvsecyiq8vgitiiprnsycw0lbrzlwbdgltc3ut5jine0yrvmco70y7okemsf2e71calcnqsrjztkjxwsl5yqmrkifvilndysz/jk3qwn68v6tz7grtlrmrnn/sd6/afew5de</latexit> What is a graph/network? node attributes Attributed graph / network edge attributes
8 8 Machine learning and the table Columns are attributes of the entity Rows are instances of an entity First Last Gender Age Income Homer Simpson M Marge Simpson F 36 0 Bart Simpson M 10 0 Lisa Simpson F 8 0 Carl Carlson M 43 Lenny Leonard M Ned Flanders M
9 9 <latexit sha1_base64="y6lmc7chuulx4krbzxquxvtmr9c=">aaacf3icbva9swnben3zm8avu0ubxsdejt6johybg8singkkiext5pilex/szqnhym+w8a/ywkjyaue/czncoykpbt6+n8popd+rqqpjffsli0vlk6uftel6xubwtr2ze6vjvhhwecxj1fczbiki8fcgheaigiw+hlo/ubz79ttqwstrdq4taiesf4lacizg6tjhlyqhzjcphucom6atcyf5qfiebuwpgqddvie29lhhljkvzwi6t9ycleiowsf+anvjnoyqizdm66brjnjomelbjyykrvrdwvia9abpamrc0o1sctiihhqls4nymyqqtttfexkltr6gvukmgfb1rdcw//oakqbn7uxesyoq8elhqsopxnsceu0kbrzl0bdgltc7ut5nine0wrznco7syfpeo6lcvnzr01k1mqdripvkgjsjs85ilvyrgveij4/kmbysn+vjerhery9p64kvz+yrp7a+fwdzakeo</latexit> Machine learning and the table Features Target First Last Gender Age Income Homer Simpson M Marge Simpson F 36 0 Predict Target Using Features Bart Simpson M 10 0 Lisa Simpson F 8 0 Carl Carlson M 43? Lenny Leonard M Ned Flanders M
10 10 <latexit sha1_base64="hge8gqzoag2bwpzjckfrnk0nltq=">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</latexit> Machine learning and the table Independence assumption: Every instance of an entity is independent of other instances: In reality instances are often connected and not independent Sex: M Income:? Age: 43 Edge type represents a relation between instances Sex: M Income: Age: 43 Sex: M Income: Age: 45 Colleague Neighbour Family
11 11 Machine Learning on Graphs Classical approach: For each node insert structural properties as additional features in the table. Typical graph features: Node degree (number of neighbours); Centrality measures (betweenness, closeness); Subgraph matches (triangle count); Many others! 2 3 1
12 12 Example: Identifying Spammers Content based spam identification: Message 50% off Win Analyse content Instant Cash Machine learning classifier Spam Not-Spam Network action based spammer identification at Tagged.com: 1 Fakhraei et al. Collective Spammer Detection in Evolving Multi-Relational Social Networks KDD 15
13 13 Example: Identifying Spammers Temporal Features Could classify 70% of spammers with 90% accuracy. 1 Fakhraei et al. Collective Spammer Detection in Evolving Multi-Relational Social Networks KDD 15
14 14 Deep learning on Graphs They are currently the state-of-the-art algorithms for many graphbased machine learning tasks.
15 15 Example: Movie recommendations Users rate movies (with a point scale of stars) User Rating 1-5 Movie Movielens dataset: Movielens 1M 6,040 Users 3,900 Movies 1,000,209 Ratings Movie features Year Genre Critic ratings User features Gender Age Occupation Harper et al. The MovieLens Datasets: History and Context. ACM Transactions on Interactive Intelligent Systems
16 16 Baseline: Collaborative Filtering Users that rate other movies similarly to you are used to predict the rating for a movie you haven t rated.
17 17 Baseline: Collaborative Filtering Users that rate other movies similarly to you are used to predict the rating for a movie you haven t rated.
18 18 Example: Movie recommendations Collaborative filtering Users that rate other movies similarly to you are used to predict the rating for a movie you haven t rated. Content-based methods These methods aim to use movie features (genre, year, critic rating etc) and user profiles (preferences, demographics) to predict which movies will be favoured by particular users. Social-based methods People like what their close friends like. Social networks can be used to improve recommendations. Hybrid methods: These aim to combined multiple previous methods using feature combination, weighted scoring, and other techniques. Konstan and Riedl. Recommender systems: from algorithms to user experience UMUAI 2012, dx.doi.org/ /s x
19 19 <latexit sha1_base64="cdbvlf1xu2cpegiklulsp13g4zo=">aaab6xicbvbns8naej34wetx1aoxxsj4koki6q3gxwnfywttkjvtpf272ytdjvbdf4ixdype/ufe/ddu2xy09cha470zzuafqedauo63s7s8srq2xtoob25t7+xw9vbvdziphj5lrkjaiduouetfccowlsqkcsiwgq6vjn7zezxmibwzoxsdmpyljzijxkq3t92hbqxq1twpyclxclkfao1u5avts1gwozrmuk3bnpuaikfkcczwxo5kglpkhrspbusljveh+ftumtm2so9eibildzmqvydygms9ikpbgvmz0pperpzpa2cmughyltpmogszrvemieni5g/s4wqzesnlkfpc3krygcrkje2nbepw5l9ejp5p7blm3zxv6/uijricwhgcgafnuidraiapdprwdk/w5gjnxxl3pmats04xcwb/4hz+aniqja0=</latexit> <latexit sha1_base64="7gvkpuuddnhfjc+fkbocycmlxkm=">aaab6xicbvbns8naej3ur1q/qh69lbbbu0leug8flx4rgltoq9lsn+3szsbstoqa+ho8efdx6j/y5r9x2+agrq8ghu/nmdmvtkuw6lrftmlldw19o7xz2dre2d2r7h88mcttjpsskyluh9rwkrt3uadk7vrzgoest8lr9drvpxjtrkluczzyikydjslbkfrp7qk36lvrbt2dgswtrya1kndsvb+6/yrlmvfijdwm47kpbjnvkjjkk0o3mzylbeqhvgopoje3qt47dujorninuajtksqz9fdetmnjxnfoo2okq7potcx/ve6g0wwqc5vmybwbl4oystah079jx2joui4touwleythq6opq5toxybglb68tpyz+lxduz2vnrpfgmu4gmm4bq8uoae30aqfgazggv7hzzhoi/pufmxbs04xcwh/4hz+anotja4=</latexit> <latexit sha1_base64="u9emzzho7buvbqhprrx9jegz6wm=">aaab7xicbvbns8naej3ur1q/qh69lbbbg5rebpvw8okxgrgfnptndtou3wzc7kqoot/ciwcvr/4fb/4bt20o2vpg4pheddpzwlqkg6777zrwvtfwn8qbla3tnd296v7bg0kyzbjpepnodkgnl0jxhwvk3k41p3eoessc3uz91hpxritqhscpd2i6ucisjkkvwrqxp56njr1qza27m5bl4hwkbgwavepxt5+wloykmatgddw3xscnggwtfflpzoanli3oghcsvttmjshn507iivx6jeq0lyvkpv6eyglszdgobwdmcwgwvan4n9fjmlokcqhsdlli80vrjgkmzpo76qvngcqxjzrpyw8lbeg1zwgtqtgqvmwxl4l/xr+ue3cxtuajskmmr3amp+dbjttgfprga4mrpmmrvdmp8+k8ox/z1pjtzbzchzifp7wbj1w=</latexit> Recommender systems with Graph ML Neural networks are now state-of-the-art for Graph ML How might we build a neural-network based recommender? User j We want to predict this rating. Rating Movie k Fully-connected neural network
20 20 <latexit sha1_base64="vt3a/twh4egbkeiuc6js1ydbvjo=">aaab73icbvbns8naej34wetx1aoxxsj4kjkion4kxjxwmlbshrdzbtptdzdhdypu0f/hxyokv/+on/+n2zyhbx0w8hhvhpl5ucqznq777swtr6yurzc2yptb2zu7lb39e51kilcfjdxrrqhrypmkvmgg01aqkbyrp81oed3xm49uazbioznkasbwt7kyewys9pau5ontgq7gyaxq1twp0clxclkfao2w8txpjiqtvbrcsdztz01nkgnlgof0xo5kmqaydhgpti2vwfad5nodx+jykl0uj8qwngiq/p7isdb6jclbkbdp63lviv7nttmtxwy5k2lmqcszrxhgkunq5hvuzyosw0ewykkyvrwrplaygjtr2ybgzb+8spyz2lxnuz2v1utfgiu4hcm4aq8uoa430aafcah4hld4c5tz4rw7h7pwjaeyoya/cd5/aeygkeq=</latexit> <latexit sha1_base64="wuadcoazdjf2hov2auxfjnqmtw4=">aaab73icbvbns8naej34wetx1aoxxsj4kjkion4kxjxwmlbshrdzbtolu5uwuxfq6k/w4khfq3/hm//gbzudtj4yelw3w8y8konmg9f9dpawv1bx1isb1c2t7z3d2t7+vu5zrahpup6qtoq15uxs3zddasdtfiui03auxe/89invmqxyzowyggg8kcxmbbsrptyfrxiqw2qc1upuw50clrkvjhuo0qprx71+snjbpseca9313mwebvageu7h1v6uayzjgge0a6neguqgmb48rsdw6am4vbakqvp190sbhdyjedlogc1qz3st8t+vm5v4miiyzhjdjzktinootiom36m+u5qypriee8xsrygmscle2iyqngrv/uvf4p81rhre7xm92sztqmahhmejehabtbibfvhaqmazvmkbo5wx5935mluuoexmafyb8/kdt6yqrg==</latexit> <latexit sha1_base64="vv4vqzhtdz+vxaal8jbzgaoaxwq=">aaab7xicbvbns8naej3ur1q/qh69lbbbg5rebpvw8okxgrgfnptndtmu3wzc7ksoot/ciwcvr/4fb/4bt20o2vpg4pheddpzwlqkg6777zrwvtfwn8qbla3tnd296v7bg0kyzbjpepnodkgnl0jxhwvk3k41p3eoessc3uz91ipxritqhscpd2i6ucisjkkvwk+9fhtmtxrvmlt3zydlxctidqo0e9wvbj9hwcwvmkmn6xhuikfonqom+atszqxpkrvrae9yqmjmtzdpzp2qe6v0szrowwrjtp09kdpymhec2s6y4taselpxp6+tyxqv5eklgxlf5ouitbjmypr30heam5rjsyjtwt5k2jbqytamvleheisvlxp/vh5d9+4uao1gkuyzjuaytsgds2jalttbbwyjeizxehns58v5dz7mrswnmdmep3a+fwbyay8r</latexit> <latexit sha1_base64="0eqrwf6ipgksd+t4xvxicwnsvyu=">aaab7xicbvbns8naej34wetx1aoxxsj4kjkion4kxjxwmlbqhrlzbtq1m03ynqg19ed48adi1f/jzx/jts1bwx8mpn6bywzemeph0hw/naxlldw19djgexnre2e3srd/b5jmm+6zrca6fvldpvdcr4gst1lnarxk3gyh1xo/+ci1eym6w1hkg5j2lygeo2il5lm3fzj1xt1k1a25u5bf4hwkcgua3cpxp5ewloykmatgtd03xscnggwtffzuzianla1pn7ctvttmjsin547jsvv6jeq0lyvkqv6eyglszcgobwdmcwdmvyn4n9fomlomcqhsdllis0vrjgkmzpi76qnngcqrjzrpyw8lbea1zwgtktsqvpmxf4l/vruqebfn1xq9skmeh3aej+dbbdthbhrga4mhpmmrvdmp8+k8ox+z1iwnmdmap3a+fwbwe48q</latexit> <latexit sha1_base64="8d2ytthki0f5v+yac4za3nqypjs=">aaab6nicbvbnswmxej3ur1q/qh69bivgqeykon4kxjxwcg2hxuo2zbahsxzjskjz+he8efdx6i/y5r8x2+5bwx8mpn6bywzelapurod9o8ra+sbmvnw7tro7t39qpzx6nemmkqtoihldjyhhgiswwg4f66aaerkj1okmt4xfewla8eq92gnkqklgiseceltifzpjqb3hnb058crxs9kaeu1b/as/tggmmbjuegn6vpfamcfaciryrnbpdesjnzar6zmqigqmzoe3zvczu4y4trqrzffc/t2re2nmveauuxi7nsteif7n9tibx4c5v2lmmaklrxemse1w8tgecs2ofvnhcnxc3yrpmghcryun5klwl19ejcff86bp3182wq0yjsqcwcmcgw9x0ii7aemafmbwdk/whir6qe/oy9faqexmmfwb+vwbnmyojq==</latexit> <latexit sha1_base64="8d2ytthki0f5v+yac4za3nqypjs=">aaab6nicbvbnswmxej3ur1q/qh69bivgqeykon4kxjxwcg2hxuo2zbahsxzjskjz+he8efdx6i/y5r8x2+5bwx8mpn6bywzelapurod9o8ra+sbmvnw7tro7t39qpzx6nemmkqtoihldjyhhgiswwg4f66aaerkj1okmt4xfewla8eq92gnkqklgiseceltifzpjqb3hnb058crxs9kaeu1b/as/tggmmbjuegn6vpfamcfaciryrnbpdesjnzar6zmqigqmzoe3zvczu4y4trqrzffc/t2re2nmveauuxi7nsteif7n9tibx4c5v2lmmaklrxemse1w8tgecs2ofvnhcnxc3yrpmghcryun5klwl19ejcff86bp3182wq0yjsqcwcmcgw9x0ii7aemafmbwdk/whir6qe/oy9faqexmmfwb+vwbnmyojq==</latexit> <latexit sha1_base64="cdbvlf1xu2cpegiklulsp13g4zo=">aaab6xicbvbns8naej34wetx1aoxxsj4koki6q3gxwnfywttkjvtpf272ytdjvbdf4ixdype/ufe/ddu2xy09cha470zzuafqedauo63s7s8srq2xtoob25t7+xw9vbvdziphj5lrkjaiduouetfccowlsqkcsiwgq6vjn7zezxmibwzoxsdmpyljzijxkq3t92hbqxq1twpyclxclkfao1u5avts1gwozrmuk3bnpuaikfkcczwxo5kglpkhrspbusljveh+ftumtm2so9eibildzmqvydygms9ikpbgvmz0pperpzpa2cmughyltpmogszrvemieni5g/s4wqzesnlkfpc3krygcrkje2nbepw5l9ejp5p7blm3zxv6/uijricwhgcgafnuidraiapdprwdk/w5gjnxxl3pmats04xcwb/4hz+aniqja0=</latexit> <latexit sha1_base64="7gvkpuuddnhfjc+fkbocycmlxkm=">aaab6xicbvbns8naej3ur1q/qh69lbbbu0leug8flx4rgltoq9lsn+3szsbstoqa+ho8efdx6j/y5r9x2+agrq8ghu/nmdmvtkuw6lrftmlldw19o7xz2dre2d2r7h88mcttjpsskyluh9rwkrt3uadk7vrzgoest8lr9drvpxjtrkluczzyikydjslbkfrp7qk36lvrbt2dgswtrya1kndsvb+6/yrlmvfijdwm47kpbjnvkjjkk0o3mzylbeqhvgopoje3qt47dujorninuajtksqz9fdetmnjxnfoo2okq7potcx/ve6g0wwqc5vmybwbl4oystah079jx2joui4touwleythq6opq5toxybglb68tpyz+lxduz2vnrpfgmu4gmm4bq8uoae30aqfgazggv7hzzhoi/pufmxbs04xcwh/4hz+anotja4=</latexit> <latexit sha1_base64="u9emzzho7buvbqhprrx9jegz6wm=">aaab7xicbvbns8naej3ur1q/qh69lbbbg5rebpvw8okxgrgfnptndtou3wzc7kqoot/ciwcvr/4fb/4bt20o2vpg4pheddpzwlqkg6777zrwvtfwn8qbla3tnd296v7bg0kyzbjpepnodkgnl0jxhwvk3k41p3eoessc3uz91hpxritqhscpd2i6ucisjkkvwrqxp56njr1qza27m5bl4hwkbgwavepxt5+wloykmatgddw3xscnggwtfflpzoanli3oghcsvttmjshn507iivx6jeq0lyvkpv6eyglszdgobwdmcwgwvan4n9fjmlokcqhsdlli80vrjgkmzpo76qvngcqxjzrpyw8lbeg1zwgtqtgqvmwxl4l/xr+ue3cxtuajskmmr3amp+dbjttgfprga4mrpmmrvdmp8+k8ox/z1pjtzbzchzifp7wbj1w=</latexit> Recommender systems with Graph ML Movies rated by user j User j Average Rating Average Movie k Fully-connected neural network Users who rated movie k
21 21 <latexit sha1_base64="nfvu6g5jsmxhjutq6tx9lkbskk0=">aaab6xicbvbns8naej3ur1q/qh69lbbbu0leug8flx4rgltoq9lsj+3szsbsbsqs+ho8efdx6j/y5r9x2+agrq8ghu/nmdmvtaxxxnw/ndlk6tr6rnmzsrw9s7tx3t940emmgposeylqh1sj4bj9w43adqqqxqhavji6nvqtr1saj/lejfmmyjqqpokmgivdpfw8xrxm1t0zydlxclkdas1e9avbt1gwozrmuk07npuaikfkcczwuulmglpkrnsahusljveh+ezuctmxsp9eibildzmpvydygms9jkpbgvmz1ivevpzp62qmugxyltpmogtzrvemieni9g/s5wqzewnlkfpc3krykcrkje2nykpwfl9ejv5z/aru3z7xgo0ijticwtgcggcx0iabaiipdabwdk/w5gjnxxl3puatjaeyoyq/cd5/ahjbjxi=</latexit> <latexit sha1_base64="jnoyvhbdtoao4sgli/giizbpodu=">aaab6xicbvbns8naej34wetx1aoxxsj4koki6q3gxznunlbqhrlzttqlm03y3ygl9cd48adi1x/kzx/jts1bwx8mpn6bywzemaqujet+o0vlk6tr66wn8ubw9s5uzw//qsezyuizrcsqfvkngkv0dtccw6lcgoccm+hwaui3h1fpnsh7m0oxiglf8ogzaqx099s96vaqbs2dgiwsrybvkndovr46vyrlmurdbnw67bmpcxkqdgccx+vopjglbej72lzu0hh1ke9phznjq/rilchb0pcp+nsip7hwozi0nte1az3vtct/vhzmoosg5zlndeo2wxrlgpietp4mpa6qgtgyhdlf7a2edaiiznh0yjyeb/7lrekf1i5r3u1ztv4v0ijbirzbcxhwdnw4hgb4wkapz/akb45wxpx352pwuuqumwfwb87nd6syjy8=</latexit> <latexit sha1_base64="parslp3scte1/zi1xf9+jnm70s8=">aaab6xicbvbns8naej3ur1q/qh69lbbbu0leug8fl16eisyw2la22027dlmjuxohhp4elx5uvpqpvplv3ly5aoudgcd7m8zmc1mpdlrut1nawv1b3yhvvra2d3b3qvshjybjnom+s2si2ye1xarffrqoetvvnmah5k1wdd31w09cg5gobxynpijpqilimipwus96t71qza27m5bl4hwkbgwavepxt5+wloykmatgddw3xscnggwtfflpzoanli3oghcsvttmjshnp07iivx6jeq0lyvkpv6eyglszdgobwdmcwgwvan4n9fjmlomcqhsdlli80vrjgkmzpo36qvngcqxjzrpyw8lbeg1zwjtqdgqvmwxl4l/vr+qe3fntuajskmmr3amp+dbbttgbprga4mbpmmrvdnsexheny95a8kpzg7hd5zph56gjys=</latexit> <latexit sha1_base64="1uklseon2r96dpcjhfzod3fknam=">aaab6xicbvbns8naej3ur1q/qh69lbbbu0leqn4kxjxwnlbqhrlzttqlm03y3qgl9cd48adi1x/kzx/jts1bwx8mpn6bywzemaqujet+o6w19y3nrfj2zwd3b/+genj0qjnmmfrzihlvcalgwsx6hhubnvqhjuob7xb8m/pbt6g0t+sdmaqyxhqoecqznva6z/pev1pz6+4czjv4balbgva/+tubjcyluromqnzdz01nkfnlobm4rfqyjsllyzrerqwsxqidfh7qljxzzucirnmshszv3xm5jbwexkhtjkkz6wvvjv7ndtmtxqu5l2lmullfoigtxcrk9jczcixmiikllclubyvsrbvlxqztssf4yy+vev+ifl337i5rzwarrhlo4btowymgnoewwuadgye8wyu8ocj5cd6dj0vryslmjuepnm8fdeynbw==</latexit> <latexit sha1_base64="8d2ytthki0f5v+yac4za3nqypjs=">aaab6nicbvbnswmxej3ur1q/qh69bivgqeykon4kxjxwcg2hxuo2zbahsxzjskjz+he8efdx6i/y5r8x2+5bwx8mpn6bywzelapurod9o8ra+sbmvnw7tro7t39qpzx6nemmkqtoihldjyhhgiswwg4f66aaerkj1okmt4xfewla8eq92gnkqklgiseceltifzpjqb3hnb058crxs9kaeu1b/as/tggmmbjuegn6vpfamcfaciryrnbpdesjnzar6zmqigqmzoe3zvczu4y4trqrzffc/t2re2nmveauuxi7nsteif7n9tibx4c5v2lmmaklrxemse1w8tgecs2ofvnhcnxc3yrpmghcryun5klwl19ejcff86bp3182wq0yjsqcwcmcgw9x0ii7aemafmbwdk/whir6qe/oy9faqexmmfwb+vwbnmyojq==</latexit> <latexit sha1_base64="8d2ytthki0f5v+yac4za3nqypjs=">aaab6nicbvbnswmxej3ur1q/qh69bivgqeykon4kxjxwcg2hxuo2zbahsxzjskjz+he8efdx6i/y5r8x2+5bwx8mpn6bywzelapurod9o8ra+sbmvnw7tro7t39qpzx6nemmkqtoihldjyhhgiswwg4f66aaerkj1okmt4xfewla8eq92gnkqklgiseceltifzpjqb3hnb058crxs9kaeu1b/as/tggmmbjuegn6vpfamcfaciryrnbpdesjnzar6zmqigqmzoe3zvczu4y4trqrzffc/t2re2nmveauuxi7nsteif7n9tibx4c5v2lmmaklrxemse1w8tgecs2ofvnhcnxc3yrpmghcryun5klwl19ejcff86bp3182wq0yjsqcwcmcgw9x0ii7aemafmbwdk/whir6qe/oy9faqexmmfwb+vwbnmyojq==</latexit> <latexit sha1_base64="cdbvlf1xu2cpegiklulsp13g4zo=">aaab6xicbvbns8naej34wetx1aoxxsj4koki6q3gxwnfywttkjvtpf272ytdjvbdf4ixdype/ufe/ddu2xy09cha470zzuafqedauo63s7s8srq2xtoob25t7+xw9vbvdziphj5lrkjaiduouetfccowlsqkcsiwgq6vjn7zezxmibwzoxsdmpyljzijxkq3t92hbqxq1twpyclxclkfao1u5avts1gwozrmuk3bnpuaikfkcczwxo5kglpkhrspbusljveh+ftumtm2so9eibildzmqvydygms9ikpbgvmz0pperpzpa2cmughyltpmogszrvemieni5g/s4wqzesnlkfpc3krygcrkje2nbepw5l9ejp5p7blm3zxv6/uijricwhgcgafnuidraiapdprwdk/w5gjnxxl3pmats04xcwb/4hz+aniqja0=</latexit> <latexit sha1_base64="7gvkpuuddnhfjc+fkbocycmlxkm=">aaab6xicbvbns8naej3ur1q/qh69lbbbu0leug8flx4rgltoq9lsn+3szsbstoqa+ho8efdx6j/y5r9x2+agrq8ghu/nmdmvtkuw6lrftmlldw19o7xz2dre2d2r7h88mcttjpsskyluh9rwkrt3uadk7vrzgoest8lr9drvpxjtrkluczzyikydjslbkfrp7qk36lvrbt2dgswtrya1kndsvb+6/yrlmvfijdwm47kpbjnvkjjkk0o3mzylbeqhvgopoje3qt47dujorninuajtksqz9fdetmnjxnfoo2okq7potcx/ve6g0wwqc5vmybwbl4oystah079jx2joui4touwleythq6opq5toxybglb68tpyz+lxduz2vnrpfgmu4gmm4bq8uoae30aqfgazggv7hzzhoi/pufmxbs04xcwh/4hz+anotja4=</latexit> <latexit sha1_base64="u9emzzho7buvbqhprrx9jegz6wm=">aaab7xicbvbns8naej3ur1q/qh69lbbbg5rebpvw8okxgrgfnptndtou3wzc7kqoot/ciwcvr/4fb/4bt20o2vpg4pheddpzwlqkg6777zrwvtfwn8qbla3tnd296v7bg0kyzbjpepnodkgnl0jxhwvk3k41p3eoessc3uz91hpxritqhscpd2i6ucisjkkvwrqxp56njr1qza27m5bl4hwkbgwavepxt5+wloykmatgddw3xscnggwtfflpzoanli3oghcsvttmjshn507iivx6jeq0lyvkpv6eyglszdgobwdmcwgwvan4n9fjmlokcqhsdlli80vrjgkmzpo76qvngcqxjzrpyw8lbeg1zwgtqtgqvmwxl4l/xr+ue3cxtuajskmmr3amp+dbjttgfprga4mrpmmrvdmp8+k8ox/z1pjtzbzchzifp7wbj1w=</latexit> Recommender systems with Graph ML Embedding Vectors Movies rated by user j All movies User j More layers Rating All users Average Movie k Fully-connected neural network Users who rated movie k
22 22 <latexit sha1_base64="nfvu6g5jsmxhjutq6tx9lkbskk0=">aaab6xicbvbns8naej3ur1q/qh69lbbbu0leug8flx4rgltoq9lsj+3szsbsbsqs+ho8efdx6j/y5r9x2+agrq8ghu/nmdmvtaxxxnw/ndlk6tr6rnmzsrw9s7tx3t940emmgposeylqh1sj4bj9w43adqqqxqhavji6nvqtr1saj/lejfmmyjqqpokmgivdpfw8xrxm1t0zydlxclkdas1e9avbt1gwozrmuk07npuaikfkcczwuulmglpkrnsahusljveh+ezuctmxsp9eibildzmpvydygms9jkpbgvmz1ivevpzp62qmugxyltpmogtzrvemieni9g/s5wqzewnlkfpc3krykcrkje2nykpwfl9ejv5z/aru3z7xgo0ijticwtgcggcx0iabaiipdabwdk/w5gjnxxl3puatjaeyoyq/cd5/ahjbjxi=</latexit> <latexit sha1_base64="jnoyvhbdtoao4sgli/giizbpodu=">aaab6xicbvbns8naej34wetx1aoxxsj4koki6q3gxznunlbqhrlzttqlm03y3ygl9cd48adi1x/kzx/jts1bwx8mpn6bywzemaqujet+o0vlk6tr66wn8ubw9s5uzw//qsezyuizrcsqfvkngkv0dtccw6lcgoccm+hwaui3h1fpnsh7m0oxiglf8ogzaqx099s96vaqbs2dgiwsrybvkndovr46vyrlmurdbnw67bmpcxkqdgccx+vopjglbej72lzu0hh1ke9phznjq/rilchb0pcp+nsip7hwozi0nte1az3vtct/vhzmoosg5zlndeo2wxrlgpietp4mpa6qgtgyhdlf7a2edaiiznh0yjyeb/7lrekf1i5r3u1ztv4v0ijbirzbcxhwdnw4hgb4wkapz/akb45wxpx352pwuuqumwfwb87nd6syjy8=</latexit> <latexit sha1_base64="parslp3scte1/zi1xf9+jnm70s8=">aaab6xicbvbns8naej3ur1q/qh69lbbbu0leug8fl16eisyw2la22027dlmjuxohhp4elx5uvpqpvplv3ly5aoudgcd7m8zmc1mpdlrut1nawv1b3yhvvra2d3b3qvshjybjnom+s2si2ye1xarffrqoetvvnmah5k1wdd31w09cg5gobxynpijpqilimipwus96t71qza27m5bl4hwkbgwavepxt5+wloykmatgddw3xscnggwtfflpzoanli3oghcsvttmjshnp07iivx6jeq0lyvkpv6eyglszdgobwdmcwgwvan4n9fjmlomcqhsdlli80vrjgkmzpo36qvngcqxjzrpyw8lbeg1zwjtqdgqvmwxl4l/vr+qe3fntuajskmmr3amp+dbbttgbprga4mbpmmrvdnsexheny95a8kpzg7hd5zph56gjys=</latexit> <latexit sha1_base64="1uklseon2r96dpcjhfzod3fknam=">aaab6xicbvbns8naej3ur1q/qh69lbbbu0leqn4kxjxwnlbqhrlzttqlm03y3qgl9cd48adi1x/kzx/jts1bwx8mpn6bywzemaqujet+o6w19y3nrfj2zwd3b/+genj0qjnmmfrzihlvcalgwsx6hhubnvqhjuob7xb8m/pbt6g0t+sdmaqyxhqoecqznva6z/pev1pz6+4czjv4balbgva/+tubjcyluromqnzdz01nkfnlobm4rfqyjsllyzrerqwsxqidfh7qljxzzucirnmshszv3xm5jbwexkhtjkkz6wvvjv7ndtmtxqu5l2lmullfoigtxcrk9jczcixmiikllclubyvsrbvlxqztssf4yy+vev+ifl337i5rzwarrhlo4btowymgnoewwuadgye8wyu8ocj5cd6dj0vryslmjuepnm8fdeynbw==</latexit> <latexit sha1_base64="8d2ytthki0f5v+yac4za3nqypjs=">aaab6nicbvbnswmxej3ur1q/qh69bivgqeykon4kxjxwcg2hxuo2zbahsxzjskjz+he8efdx6i/y5r8x2+5bwx8mpn6bywzelapurod9o8ra+sbmvnw7tro7t39qpzx6nemmkqtoihldjyhhgiswwg4f66aaerkj1okmt4xfewla8eq92gnkqklgiseceltifzpjqb3hnb058crxs9kaeu1b/as/tggmmbjuegn6vpfamcfaciryrnbpdesjnzar6zmqigqmzoe3zvczu4y4trqrzffc/t2re2nmveauuxi7nsteif7n9tibx4c5v2lmmaklrxemse1w8tgecs2ofvnhcnxc3yrpmghcryun5klwl19ejcff86bp3182wq0yjsqcwcmcgw9x0ii7aemafmbwdk/whir6qe/oy9faqexmmfwb+vwbnmyojq==</latexit> <latexit sha1_base64="8d2ytthki0f5v+yac4za3nqypjs=">aaab6nicbvbnswmxej3ur1q/qh69bivgqeykon4kxjxwcg2hxuo2zbahsxzjskjz+he8efdx6i/y5r8x2+5bwx8mpn6bywzelapurod9o8ra+sbmvnw7tro7t39qpzx6nemmkqtoihldjyhhgiswwg4f66aaerkj1okmt4xfewla8eq92gnkqklgiseceltifzpjqb3hnb058crxs9kaeu1b/as/tggmmbjuegn6vpfamcfaciryrnbpdesjnzar6zmqigqmzoe3zvczu4y4trqrzffc/t2re2nmveauuxi7nsteif7n9tibx4c5v2lmmaklrxemse1w8tgecs2ofvnhcnxc3yrpmghcryun5klwl19ejcff86bp3182wq0yjsqcwcmcgw9x0ii7aemafmbwdk/whir6qe/oy9faqexmmfwb+vwbnmyojq==</latexit> <latexit sha1_base64="cdbvlf1xu2cpegiklulsp13g4zo=">aaab6xicbvbns8naej34wetx1aoxxsj4koki6q3gxwnfywttkjvtpf272ytdjvbdf4ixdype/ufe/ddu2xy09cha470zzuafqedauo63s7s8srq2xtoob25t7+xw9vbvdziphj5lrkjaiduouetfccowlsqkcsiwgq6vjn7zezxmibwzoxsdmpyljzijxkq3t92hbqxq1twpyclxclkfao1u5avts1gwozrmuk3bnpuaikfkcczwxo5kglpkhrspbusljveh+ftumtm2so9eibildzmqvydygms9ikpbgvmz0pperpzpa2cmughyltpmogszrvemieni5g/s4wqzesnlkfpc3krygcrkje2nbepw5l9ejp5p7blm3zxv6/uijricwhgcgafnuidraiapdprwdk/w5gjnxxl3pmats04xcwb/4hz+aniqja0=</latexit> <latexit sha1_base64="7gvkpuuddnhfjc+fkbocycmlxkm=">aaab6xicbvbns8naej3ur1q/qh69lbbbu0leug8flx4rgltoq9lsn+3szsbstoqa+ho8efdx6j/y5r9x2+agrq8ghu/nmdmvtkuw6lrftmlldw19o7xz2dre2d2r7h88mcttjpsskyluh9rwkrt3uadk7vrzgoest8lr9drvpxjtrkluczzyikydjslbkfrp7qk36lvrbt2dgswtrya1kndsvb+6/yrlmvfijdwm47kpbjnvkjjkk0o3mzylbeqhvgopoje3qt47dujorninuajtksqz9fdetmnjxnfoo2okq7potcx/ve6g0wwqc5vmybwbl4oystah079jx2joui4touwleythq6opq5toxybglb68tpyz+lxduz2vnrpfgmu4gmm4bq8uoae30aqfgazggv7hzzhoi/pufmxbs04xcwh/4hz+anotja4=</latexit> <latexit sha1_base64="u9emzzho7buvbqhprrx9jegz6wm=">aaab7xicbvbns8naej3ur1q/qh69lbbbg5rebpvw8okxgrgfnptndtou3wzc7kqoot/ciwcvr/4fb/4bt20o2vpg4pheddpzwlqkg6777zrwvtfwn8qbla3tnd296v7bg0kyzbjpepnodkgnl0jxhwvk3k41p3eoessc3uz91hpxritqhscpd2i6ucisjkkvwrqxp56njr1qza27m5bl4hwkbgwavepxt5+wloykmatgddw3xscnggwtfflpzoanli3oghcsvttmjshn507iivx6jeq0lyvkpv6eyglszdgobwdmcwgwvan4n9fjmlokcqhsdlli80vrjgkmzpo76qvngcqxjzrpyw8lbeg1zwgtqtgqvmwxl4l/xr+ue3cxtuajskmmr3amp+dbjttgfprga4mrpmmrvdmp8+k8ox/z1pjtzbzchzifp7wbj1w=</latexit> Recommender systems with Graph ML Embedding Vectors Back-propagation Movies rated by user j All movies User j More layers Rating All users Average Movie k Fully-connected neural network Users who rated movie k
23 23 Graph convolutional neural networks This system is a graph convolutional neural network. They are general architectures can be used to predict node or edge attributes in arbitrary graphs, not only for recommender systems. Labelled Examples Node features or Node Embeddings X9 X 6 X 8 X 7 X 0 X 10 X 2 X 1 Hidden Layers Z9 Z 6 Z 8 Z 7 Z 0 Z 10 Z 2 Z 1 Y 8 Y 10 Y 2 C X 5 X 4 X 3 F Z 5 Z 4 Z 3 Hamilton et al. Inductive Representation Learning on Large Graphs, NIPS 2017 Kipf and Welling. Semi-supervised classification with graph convolutional networks. ICLR 2017.
24 24 <latexit sha1_base64="gxn56ljn7lqqgtpdsbssxbgfx9i=">aaacjxicbvdnsgmxgmzwv1r/qh69bivgxbirgnoqcl48sqvrc92yznnsg5rnlsm3ytnu03jxvbx4qcj48lvm2z1o60bgmjkvytd+llgg2/6yckvlk6trxfxsxubw9k55d+9br4mireejeamwtzqtxligcbcsfstgql+wpj+4nvjnr6y0j+q9dgpwculp8obtakbyylduoahnnsy9zbcrk9bleyzh2ax2bkkykdnlpi5pcsvc3uv0mmvukveu2fv7crxinjxuui66vx673ygmiznabdg67dgxdfkigfpbspkbabytoia91jzukpdptjpdm8nhruniiflmsmbt9fdeskkth6fvkigbvp73juj/xjub4kktchknwcsdprqkakoej53hllemghgaqqji5q+y9onpduyzjvocm7/yimmcvi+rzt1zpvbl2yiia3sijpgdzlen3aa6aickntergqn368v6sz6sz1m0youz++gpro8fajwm9a==</latexit> <latexit sha1_base64="d1jbitse5hz0pwbyxninozlywog=">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</latexit> Recommender performance How can we measure performance of predicting movie ratings? Mean average error (MAE): Root mean square error (RMSE): Caveats: A lower error between predictions and actual ratings does not mean a method performs well in generating recommendations! To be actually useful recommendations should be serendipitous, diverse, explainable, use context such as time of day and previous queries The user experience is critical! Sean M. McNee et al. Being Accurate is Not Enough: How Accuracy Metrics have hurt Recommender Systems CHI 2006,
25 25 Example: Movie recommendations Traditional Collaborative Filtering (Nearest Neighbour) Collaborative Filtering (Matrix Factorization) MAE RMSE Scalable Prediction Online Training Graph ML MAE RMSE Scalable Prediction Online Training Graph Conv Nets
26 26 Example: Movie recommendations Neural network-based Graph ML methods allow: Leveraging high-performance ML packages such as Tensorflow. Scalable training and prediction to billions of users and items. Integration of ratings Users Movies Rating
27 27 Example: Movie recommendations Neural network-based Graph ML methods allow: Leveraging high-performance ML packages such as tensorflow Scalable training and inference to billions of users and items. Integration of ratings, user and movie information Users Attributes Movies Attributes Rating Attributes Attributes
28 28 Example: Movie recommendations Neural network-based Graph ML methods allow: Leveraging high-performance ML packages such as Tensorflow Scalable training and inference to billions of users and items. Integration of ratings, user and movie information, and social network connections in a single system! Users Attributes Movies Attributes Rating Social Connections Attributes Attributes
29 29 The cold start problem With no previous ratings, collaborative filtering cannot provide sensible recommendations for a new user or movie. User Movies Rating?
30 30 <latexit sha1_base64="nfvu6g5jsmxhjutq6tx9lkbskk0=">aaab6xicbvbns8naej3ur1q/qh69lbbbu0leug8flx4rgltoq9lsj+3szsbsbsqs+ho8efdx6j/y5r9x2+agrq8ghu/nmdmvtaxxxnw/ndlk6tr6rnmzsrw9s7tx3t940emmgposeylqh1sj4bj9w43adqqqxqhavji6nvqtr1saj/lejfmmyjqqpokmgivdpfw8xrxm1t0zydlxclkdas1e9avbt1gwozrmuk07npuaikfkcczwuulmglpkrnsahusljveh+ezuctmxsp9eibildzmpvydygms9jkpbgvmz1ivevpzp62qmugxyltpmogtzrvemieni9g/s5wqzewnlkfpc3krykcrkje2nykpwfl9ejv5z/aru3z7xgo0ijticwtgcggcx0iabaiipdabwdk/w5gjnxxl3puatjaeyoyq/cd5/ahjbjxi=</latexit> <latexit sha1_base64="jnoyvhbdtoao4sgli/giizbpodu=">aaab6xicbvbns8naej34wetx1aoxxsj4koki6q3gxznunlbqhrlzttqlm03y3ygl9cd48adi1x/kzx/jts1bwx8mpn6bywzemaqujet+o0vlk6tr66wn8ubw9s5uzw//qsezyuizrcsqfvkngkv0dtccw6lcgoccm+hwaui3h1fpnsh7m0oxiglf8ogzaqx099s96vaqbs2dgiwsrybvkndovr46vyrlmurdbnw67bmpcxkqdgccx+vopjglbej72lzu0hh1ke9phznjq/rilchb0pcp+nsip7hwozi0nte1az3vtct/vhzmoosg5zlndeo2wxrlgpietp4mpa6qgtgyhdlf7a2edaiiznh0yjyeb/7lrekf1i5r3u1ztv4v0ijbirzbcxhwdnw4hgb4wkapz/akb45wxpx352pwuuqumwfwb87nd6syjy8=</latexit> <latexit sha1_base64="parslp3scte1/zi1xf9+jnm70s8=">aaab6xicbvbns8naej3ur1q/qh69lbbbu0leug8fl16eisyw2la22027dlmjuxohhp4elx5uvpqpvplv3ly5aoudgcd7m8zmc1mpdlrut1nawv1b3yhvvra2d3b3qvshjybjnom+s2si2ye1xarffrqoetvvnmah5k1wdd31w09cg5gobxynpijpqilimipwus96t71qza27m5bl4hwkbgwavepxt5+wloykmatgddw3xscnggwtfflpzoanli3oghcsvttmjshnp07iivx6jeq0lyvkpv6eyglszdgobwdmcwgwvan4n9fjmlomcqhsdlli80vrjgkmzpo36qvngcqxjzrpyw8lbeg1zwjtqdgqvmwxl4l/vr+qe3fntuajskmmr3amp+dbbttgbprga4mbpmmrvdnsexheny95a8kpzg7hd5zph56gjys=</latexit> <latexit sha1_base64="1uklseon2r96dpcjhfzod3fknam=">aaab6xicbvbns8naej3ur1q/qh69lbbbu0leqn4kxjxwnlbqhrlzttqlm03y3qgl9cd48adi1x/kzx/jts1bwx8mpn6bywzemaqujet+o6w19y3nrfj2zwd3b/+genj0qjnmmfrzihlvcalgwsx6hhubnvqhjuob7xb8m/pbt6g0t+sdmaqyxhqoecqznva6z/pev1pz6+4czjv4balbgva/+tubjcyluromqnzdz01nkfnlobm4rfqyjsllyzrerqwsxqidfh7qljxzzucirnmshszv3xm5jbwexkhtjkkz6wvvjv7ndtmtxqu5l2lmullfoigtxcrk9jczcixmiikllclubyvsrbvlxqztssf4yy+vev+ifl337i5rzwarrhlo4btowymgnoewwuadgye8wyu8ocj5cd6dj0vryslmjuepnm8fdeynbw==</latexit> <latexit sha1_base64="8d2ytthki0f5v+yac4za3nqypjs=">aaab6nicbvbnswmxej3ur1q/qh69bivgqeykon4kxjxwcg2hxuo2zbahsxzjskjz+he8efdx6i/y5r8x2+5bwx8mpn6bywzelapurod9o8ra+sbmvnw7tro7t39qpzx6nemmkqtoihldjyhhgiswwg4f66aaerkj1okmt4xfewla8eq92gnkqklgiseceltifzpjqb3hnb058crxs9kaeu1b/as/tggmmbjuegn6vpfamcfaciryrnbpdesjnzar6zmqigqmzoe3zvczu4y4trqrzffc/t2re2nmveauuxi7nsteif7n9tibx4c5v2lmmaklrxemse1w8tgecs2ofvnhcnxc3yrpmghcryun5klwl19ejcff86bp3182wq0yjsqcwcmcgw9x0ii7aemafmbwdk/whir6qe/oy9faqexmmfwb+vwbnmyojq==</latexit> <latexit sha1_base64="8d2ytthki0f5v+yac4za3nqypjs=">aaab6nicbvbnswmxej3ur1q/qh69bivgqeykon4kxjxwcg2hxuo2zbahsxzjskjz+he8efdx6i/y5r8x2+5bwx8mpn6bywzelapurod9o8ra+sbmvnw7tro7t39qpzx6nemmkqtoihldjyhhgiswwg4f66aaerkj1okmt4xfewla8eq92gnkqklgiseceltifzpjqb3hnb058crxs9kaeu1b/as/tggmmbjuegn6vpfamcfaciryrnbpdesjnzar6zmqigqmzoe3zvczu4y4trqrzffc/t2re2nmveauuxi7nsteif7n9tibx4c5v2lmmaklrxemse1w8tgecs2ofvnhcnxc3yrpmghcryun5klwl19ejcff86bp3182wq0yjsqcwcmcgw9x0ii7aemafmbwdk/whir6qe/oy9faqexmmfwb+vwbnmyojq==</latexit> <latexit sha1_base64="cdbvlf1xu2cpegiklulsp13g4zo=">aaab6xicbvbns8naej34wetx1aoxxsj4koki6q3gxwnfywttkjvtpf272ytdjvbdf4ixdype/ufe/ddu2xy09cha470zzuafqedauo63s7s8srq2xtoob25t7+xw9vbvdziphj5lrkjaiduouetfccowlsqkcsiwgq6vjn7zezxmibwzoxsdmpyljzijxkq3t92hbqxq1twpyclxclkfao1u5avts1gwozrmuk3bnpuaikfkcczwxo5kglpkhrspbusljveh+ftumtm2so9eibildzmqvydygms9ikpbgvmz0pperpzpa2cmughyltpmogszrvemieni5g/s4wqzesnlkfpc3krygcrkje2nbepw5l9ejp5p7blm3zxv6/uijricwhgcgafnuidraiapdprwdk/w5gjnxxl3pmats04xcwb/4hz+aniqja0=</latexit> <latexit sha1_base64="7gvkpuuddnhfjc+fkbocycmlxkm=">aaab6xicbvbns8naej3ur1q/qh69lbbbu0leug8flx4rgltoq9lsn+3szsbstoqa+ho8efdx6j/y5r9x2+agrq8ghu/nmdmvtkuw6lrftmlldw19o7xz2dre2d2r7h88mcttjpsskyluh9rwkrt3uadk7vrzgoest8lr9drvpxjtrkluczzyikydjslbkfrp7qk36lvrbt2dgswtrya1kndsvb+6/yrlmvfijdwm47kpbjnvkjjkk0o3mzylbeqhvgopoje3qt47dujorninuajtksqz9fdetmnjxnfoo2okq7potcx/ve6g0wwqc5vmybwbl4oystah079jx2joui4touwleythq6opq5toxybglb68tpyz+lxduz2vnrpfgmu4gmm4bq8uoae30aqfgazggv7hzzhoi/pufmxbs04xcwh/4hz+anotja4=</latexit> <latexit sha1_base64="u9emzzho7buvbqhprrx9jegz6wm=">aaab7xicbvbns8naej3ur1q/qh69lbbbg5rebpvw8okxgrgfnptndtou3wzc7kqoot/ciwcvr/4fb/4bt20o2vpg4pheddpzwlqkg6777zrwvtfwn8qbla3tnd296v7bg0kyzbjpepnodkgnl0jxhwvk3k41p3eoessc3uz91hpxritqhscpd2i6ucisjkkvwrqxp56njr1qza27m5bl4hwkbgwavepxt5+wloykmatgddw3xscnggwtfflpzoanli3oghcsvttmjshn507iivx6jeq0lyvkpv6eyglszdgobwdmcwgwvan4n9fjmlokcqhsdlli80vrjgkmzpo76qvngcqxjzrpyw8lbeg1zwgtqtgqvmwxl4l/xr+ue3cxtuajskmmr3amp+dbjttgfprga4mrpmmrvdmp8+k8ox/z1pjtzbzchzifp7wbj1w=</latexit> Recommender systems with Graph ML Embedding Vectors Back-propagation Movies rated by user j All movies User j More layers Rating All users Average Movie k Fully-connected neural network Users who rated movie k
31 31 <latexit sha1_base64="rcegekgpro8h/yrziprx88ds5gw=">aaab6xicbvbns8naej3ur1q/qh69lbbbu0leug8fl56korgfnptndtmu3wzc7kqoot/biwcvr/4jb/4bt20o2vpg4pheddpzwlqkg6777zrwvtfwn8qbla3tnd296v7bo0kyzbjpepnodkgnl0jxhwvk3k41p3eoesscxu/91hpxritqaccpd2i6ucisjkkv7qpeba9ac+vudgszeawpqyfmr/rv7scsi7lcjqkxhc9nmciprsekn1s6meepzsm64b1lfy25cflzqrnyypu+irjtsygzqb8nchobm45d2xlthjpfbyr+53uyjc6dxkg0q67yffgusyijmf5n+kjzhnjscwva2fsjg1jngdp0kjyeb/hlzekf1a/q3t15rdeo0ijderzdkxhwaq24gsb4wgaaz/akb450xpx352pewnkkmup4a+fzb4lyjx0=</latexit> <latexit sha1_base64="parslp3scte1/zi1xf9+jnm70s8=">aaab6xicbvbns8naej3ur1q/qh69lbbbu0leug8fl16eisyw2la22027dlmjuxohhp4elx5uvpqpvplv3ly5aoudgcd7m8zmc1mpdlrut1nawv1b3yhvvra2d3b3qvshjybjnom+s2si2ye1xarffrqoetvvnmah5k1wdd31w09cg5gobxynpijpqilimipwus96t71qza27m5bl4hwkbgwavepxt5+wloykmatgddw3xscnggwtfflpzoanli3oghcsvttmjshnp07iivx6jeq0lyvkpv6eyglszdgobwdmcwgwvan4n9fjmlomcqhsdlli80vrjgkmzpo36qvngcqxjzrpyw8lbeg1zwjtqdgqvmwxl4l/vr+qe3fntuajskmmr3amp+dbbttgbprga4mbpmmrvdnsexheny95a8kpzg7hd5zph56gjys=</latexit> <latexit sha1_base64="9rlutxdm2qhb72hiu61llonsqz0=">aaab6xicbvbns8naej3ur1q/qh69lbbbu0leqn4kxjxwnlbqhrlzbtqlm03ynqgl9cd48adi1x/kzx/jts1bwx8mpn6bywzemeph0hw/ndla+sbmvnm7sro7t39qptx6nemmgfdzihpdcanhuijuo0djo6nmna4lb4fjm5nffulaieq94ctlquyhskscubtsfdt3+twaw3fnikvek0gncrt61a/eigfzzbuysy3pem6kqu41cib5tnllde8pg9mh71qqamxnkm9pnzizqwxilghbcslc/t2r09iysrzazpjiycx7m/e/r5thdbxkqquzcsuwi6jmekzi7g8yejozlbnlknpc3kryigrk0kztssf4yy+vev+ifl337i5rzwarrhlo4btowymgnoewwuadgye8wyu8odj5cd6dj0vryslmjuepnm8fxygnya==</latexit> <latexit sha1_base64="9dh+jvp2g9br00tuyokv9c79kvo=">aaab6xicbvbns8naej3ur1q/qh69lbbbu0leug8flx4rgltoq9lsj+nszsbsboqs+ho8efdx6j/y5r9x2+ag1qcdj/dmmjkxzojr47pftmvldw19o7pz29re2d2r7x886drxdh2wilr1q6prcim+4uzgn1nik1bgjxxfz/zoiyrnu3lvjhkgcy0ljzijxkp38cab1btu052d/cvesrpqoj2of/ahkcstliyjqnxpczmtffqzzgroa/1cy0bzmmbys1tsbhvqze+dkhordemuklvskln6c6kgidatjlsdctujveznxp+8xm6iy6dgmssnsrzyfowcmjtm/izdrpazmbgemsxtrysnqklm2hrqngrv+ew/xd9rxjw92/ngq1wmuyujoizt8oacwnadbfcbqqxp8akvjncentfnfdfaccqzq/gf5+mbxwanyq==</latexit> <latexit sha1_base64="8d2ytthki0f5v+yac4za3nqypjs=">aaab6nicbvbnswmxej3ur1q/qh69bivgqeykon4kxjxwcg2hxuo2zbahsxzjskjz+he8efdx6i/y5r8x2+5bwx8mpn6bywzelapurod9o8ra+sbmvnw7tro7t39qpzx6nemmkqtoihldjyhhgiswwg4f66aaerkj1okmt4xfewla8eq92gnkqklgiseceltifzpjqb3hnb058crxs9kaeu1b/as/tggmmbjuegn6vpfamcfaciryrnbpdesjnzar6zmqigqmzoe3zvczu4y4trqrzffc/t2re2nmveauuxi7nsteif7n9tibx4c5v2lmmaklrxemse1w8tgecs2ofvnhcnxc3yrpmghcryun5klwl19ejcff86bp3182wq0yjsqcwcmcgw9x0ii7aemafmbwdk/whir6qe/oy9faqexmmfwb+vwbnmyojq==</latexit> <latexit sha1_base64="8d2ytthki0f5v+yac4za3nqypjs=">aaab6nicbvbnswmxej3ur1q/qh69bivgqeykon4kxjxwcg2hxuo2zbahsxzjskjz+he8efdx6i/y5r8x2+5bwx8mpn6bywzelapurod9o8ra+sbmvnw7tro7t39qpzx6nemmkqtoihldjyhhgiswwg4f66aaerkj1okmt4xfewla8eq92gnkqklgiseceltifzpjqb3hnb058crxs9kaeu1b/as/tggmmbjuegn6vpfamcfaciryrnbpdesjnzar6zmqigqmzoe3zvczu4y4trqrzffc/t2re2nmveauuxi7nsteif7n9tibx4c5v2lmmaklrxemse1w8tgecs2ofvnhcnxc3yrpmghcryun5klwl19ejcff86bp3182wq0yjsqcwcmcgw9x0ii7aemafmbwdk/whir6qe/oy9faqexmmfwb+vwbnmyojq==</latexit> <latexit sha1_base64="cdbvlf1xu2cpegiklulsp13g4zo=">aaab6xicbvbns8naej34wetx1aoxxsj4koki6q3gxwnfywttkjvtpf272ytdjvbdf4ixdype/ufe/ddu2xy09cha470zzuafqedauo63s7s8srq2xtoob25t7+xw9vbvdziphj5lrkjaiduouetfccowlsqkcsiwgq6vjn7zezxmibwzoxsdmpyljzijxkq3t92hbqxq1twpyclxclkfao1u5avts1gwozrmuk3bnpuaikfkcczwxo5kglpkhrspbusljveh+ftumtm2so9eibildzmqvydygms9ikpbgvmz0pperpzpa2cmughyltpmogszrvemieni5g/s4wqzesnlkfpc3krygcrkje2nbepw5l9ejp5p7blm3zxv6/uijricwhgcgafnuidraiapdprwdk/w5gjnxxl3pmats04xcwb/4hz+aniqja0=</latexit> <latexit sha1_base64="7gvkpuuddnhfjc+fkbocycmlxkm=">aaab6xicbvbns8naej3ur1q/qh69lbbbu0leug8flx4rgltoq9lsn+3szsbstoqa+ho8efdx6j/y5r9x2+agrq8ghu/nmdmvtkuw6lrftmlldw19o7xz2dre2d2r7h88mcttjpsskyluh9rwkrt3uadk7vrzgoest8lr9drvpxjtrkluczzyikydjslbkfrp7qk36lvrbt2dgswtrya1kndsvb+6/yrlmvfijdwm47kpbjnvkjjkk0o3mzylbeqhvgopoje3qt47dujorninuajtksqz9fdetmnjxnfoo2okq7potcx/ve6g0wwqc5vmybwbl4oystah079jx2joui4touwleythq6opq5toxybglb68tpyz+lxduz2vnrpfgmu4gmm4bq8uoae30aqfgazggv7hzzhoi/pufmxbs04xcwh/4hz+anotja4=</latexit> <latexit sha1_base64="u9emzzho7buvbqhprrx9jegz6wm=">aaab7xicbvbns8naej3ur1q/qh69lbbbg5rebpvw8okxgrgfnptndtou3wzc7kqoot/ciwcvr/4fb/4bt20o2vpg4pheddpzwlqkg6777zrwvtfwn8qbla3tnd296v7bg0kyzbjpepnodkgnl0jxhwvk3k41p3eoessc3uz91hpxritqhscpd2i6ucisjkkvwrqxp56njr1qza27m5bl4hwkbgwavepxt5+wloykmatgddw3xscnggwtfflpzoanli3oghcsvttmjshn507iivx6jeq0lyvkpv6eyglszdgobwdmcwgwvan4n9fjmlokcqhsdlli80vrjgkmzpo76qvngcqxjzrpyw8lbeg1zwgtqtgqvmwxl4l/xr+ue3cxtuajskmmr3amp+dbjttgfprga4mrpmmrvdmp8+k8ox/z1pjtzbzchzifp7wbj1w=</latexit> Recommender systems with Graph ML Back-propagation Input Features Movies rated by user j Movie Features User j More layers Rating User Features Average Movie k Fully-connected neural network Users who rated movie k
32 32 The cold start problem Extreme example: two separated groups of users and movies. User Attributes Movies Attributes Rating Train Attributes Attributes Attributes Attributes Test Attributes Attributes
33 33 Movie recommendations: Cold Start Traditional MAE RMSE Content Based Collaborative Filtering (Nearest Neighbour) Collaborative Filtering (Matrix Factorization) Graph ML MAE RMSE Graph Conv Net
34 34 Recommender systems in practice
35 35 Conclusions General-purpose algorithms for machine learning on graphs are becoming competitive with niche and purpose built methods Built on deep learning technology Support online and batch training Scalable to billions of nodes Support multiple node and edge types Can transfer knowledge from one graph to another Can use attributes of entities Many problems living in tables are moving to graphs Drug interactions Recommender systems Social networking Spam detection Entity resolution Fraud detection Biomedical interactions
36 Thank you!
Parallel learning of content recommendations using map- reduce
Parallel learning of content recommendations using map- reduce Michael Percy Stanford University Abstract In this paper, machine learning within the map- reduce paradigm for ranking
More informationCS224W Project: Recommendation System Models in Product Rating Predictions
CS224W Project: Recommendation System Models in Product Rating Predictions Xiaoye Liu xiaoye@stanford.edu Abstract A product recommender system based on product-review information and metadata history
More informationDeep Learning on Graphs
Deep Learning on Graphs with Graph Convolutional Networks Hidden layer Hidden layer Input Output ReLU ReLU, 22 March 2017 joint work with Max Welling (University of Amsterdam) BDL Workshop @ NIPS 2016
More informationCollaborative Filtering using Euclidean Distance in Recommendation Engine
Indian Journal of Science and Technology, Vol 9(37), DOI: 10.17485/ijst/2016/v9i37/102074, October 2016 ISSN (Print) : 0974-6846 ISSN (Online) : 0974-5645 Collaborative Filtering using Euclidean Distance
More informationCS249: ADVANCED DATA MINING
CS249: ADVANCED DATA MINING Recommender Systems II Instructor: Yizhou Sun yzsun@cs.ucla.edu May 31, 2017 Recommender Systems Recommendation via Information Network Analysis Hybrid Collaborative Filtering
More informationECS289: Scalable Machine Learning
ECS289: Scalable Machine Learning Cho-Jui Hsieh UC Davis Sept 24, 2015 Course Information Website: www.stat.ucdavis.edu/~chohsieh/ecs289g_scalableml.html My office: Mathematical Sciences Building (MSB)
More informationCast a Net over Your Data Lake Harnessing the power of graph analytics
Cast a Net over Your Data Lake Harnessing the power of graph analytics Natalia Rümmele Yow!Data 2017 www.data61.csiro.au Graph Analytics: Intro Graph or Network Nodes (vertices) Links (edges) Examples
More informationProject Report. An Introduction to Collaborative Filtering
Project Report An Introduction to Collaborative Filtering Siobhán Grayson 12254530 COMP30030 School of Computer Science and Informatics College of Engineering, Mathematical & Physical Sciences University
More informationHybrid Recommendation System Using Clustering and Collaborative Filtering
Hybrid Recommendation System Using Clustering and Collaborative Filtering Roshni Padate Assistant Professor roshni@frcrce.ac.in Priyanka Bane B.E. Student priyankabane56@gmail.com Jayesh Kudase B.E. Student
More informationA Scalable, Accurate Hybrid Recommender System
A Scalable, Accurate Hybrid Recommender System Mustansar Ali Ghazanfar and Adam Prugel-Bennett School of Electronics and Computer Science University of Southampton Highfield Campus, SO17 1BJ, United Kingdom
More informationDeep Learning for Recommender Systems
join at Slido.com with #bigdata2018 Deep Learning for Recommender Systems Oliver Gindele @tinyoli oliver.gindele@datatonic.com Big Data Conference Vilnius 28.11.2018 Who is Oliver? + Head of Machine Learning
More informationThe Principle and Improvement of the Algorithm of Matrix Factorization Model based on ALS
of the Algorithm of Matrix Factorization Model based on ALS 12 Yunnan University, Kunming, 65000, China E-mail: shaw.xuan820@gmail.com Chao Yi 3 Yunnan University, Kunming, 65000, China E-mail: yichao@ynu.edu.cn
More informationBatch-Incremental vs. Instance-Incremental Learning in Dynamic and Evolving Data
Batch-Incremental vs. Instance-Incremental Learning in Dynamic and Evolving Data Jesse Read 1, Albert Bifet 2, Bernhard Pfahringer 2, Geoff Holmes 2 1 Department of Signal Theory and Communications Universidad
More informationDeep Learning on Graphs
Deep Learning on Graphs with Graph Convolutional Networks Hidden layer Hidden layer Input Output ReLU ReLU, 6 April 2017 joint work with Max Welling (University of Amsterdam) The success story of deep
More informationCOMP 465: Data Mining Recommender Systems
//0 movies COMP 6: Data Mining Recommender Systems Slides Adapted From: www.mmds.org (Mining Massive Datasets) movies Compare predictions with known ratings (test set T)????? Test Data Set Root-mean-square
More informationECS289: Scalable Machine Learning
ECS289: Scalable Machine Learning Cho-Jui Hsieh UC Davis Sept 22, 2016 Course Information Website: http://www.stat.ucdavis.edu/~chohsieh/teaching/ ECS289G_Fall2016/main.html My office: Mathematical Sciences
More informationRecommender Systems. Nivio Ziviani. Junho de Departamento de Ciência da Computação da UFMG
Recommender Systems Nivio Ziviani Departamento de Ciência da Computação da UFMG Junho de 2012 1 Introduction Chapter 1 of Recommender Systems Handbook Ricci, Rokach, Shapira and Kantor (editors), 2011.
More informationChapter 1. Social Media and Social Computing. October 2012 Youn-Hee Han
Chapter 1. Social Media and Social Computing October 2012 Youn-Hee Han http://link.koreatech.ac.kr 1.1 Social Media A rapid development and change of the Web and the Internet Participatory web application
More informationMachine Learning using MapReduce
Machine Learning using MapReduce What is Machine Learning Machine learning is a subfield of artificial intelligence concerned with techniques that allow computers to improve their outputs based on previous
More informationData Mining Techniques
Data Mining Techniques CS 60 - Section - Fall 06 Lecture Jan-Willem van de Meent (credit: Andrew Ng, Alex Smola, Yehuda Koren, Stanford CS6) Recommender Systems The Long Tail (from: https://www.wired.com/00/0/tail/)
More informationD B M G Data Base and Data Mining Group of Politecnico di Torino
DataBase and Data Mining Group of Data mining fundamentals Data Base and Data Mining Group of Data analysis Most companies own huge databases containing operational data textual documents experiment results
More informationDeep Character-Level Click-Through Rate Prediction for Sponsored Search
Deep Character-Level Click-Through Rate Prediction for Sponsored Search Bora Edizel - Phd Student UPF Amin Mantrach - Criteo Research Xiao Bai - Oath This work was done at Yahoo and will be presented as
More informationTensorFlow: A System for Learning-Scale Machine Learning. Google Brain
TensorFlow: A System for Learning-Scale Machine Learning Google Brain The Problem Machine learning is everywhere This is in large part due to: 1. Invention of more sophisticated machine learning models
More informationPart I: Data Mining Foundations
Table of Contents 1. Introduction 1 1.1. What is the World Wide Web? 1 1.2. A Brief History of the Web and the Internet 2 1.3. Web Data Mining 4 1.3.1. What is Data Mining? 6 1.3.2. What is Web Mining?
More informationBig Data Analytics Influx of data pertaining to the 4Vs, i.e. Volume, Veracity, Velocity and Variety
Holistic Analysis of Multi-Source, Multi- Feature Data: Modeling and Computation Challenges Big Data Analytics Influx of data pertaining to the 4Vs, i.e. Volume, Veracity, Velocity and Variety Abhishek
More informationHolistic Analysis of Multi-Source, Multi- Feature Data: Modeling and Computation Challenges
Holistic Analysis of Multi-Source, Multi- Feature Data: Modeling and Computation Challenges Abhishek Santra 1 and Sanjukta Bhowmick 2 1 Information Technology Laboratory, CSE Department, University of
More informationUsing Data Mining to Determine User-Specific Movie Ratings
Available Online at www.ijcsmc.com International Journal of Computer Science and Mobile Computing A Monthly Journal of Computer Science and Information Technology ISSN 2320 088X IMPACT FACTOR: 6.017 IJCSMC,
More informationCPSC 340: Machine Learning and Data Mining. Recommender Systems Fall 2017
CPSC 340: Machine Learning and Data Mining Recommender Systems Fall 2017 Assignment 4: Admin Due tonight, 1 late day for Monday, 2 late days for Wednesday. Assignment 5: Posted, due Monday of last week
More informationOrange3 Data Fusion Documentation. Biolab
Biolab Mar 07, 2018 Widgets 1 IMDb Actors 1 2 Chaining 5 3 Completion Scoring 9 4 Fusion Graph 13 5 Latent Factors 17 6 Matrix Sampler 21 7 Mean Fuser 25 8 Movie Genres 29 9 Movie Ratings 33 10 Table
More informationCSE 158 Lecture 8. Web Mining and Recommender Systems. Extensions of latent-factor models, (and more on the Netflix prize)
CSE 158 Lecture 8 Web Mining and Recommender Systems Extensions of latent-factor models, (and more on the Netflix prize) Summary so far Recap 1. Measuring similarity between users/items for binary prediction
More informationThis Talk. 1) Node embeddings. Map nodes to low-dimensional embeddings. 2) Graph neural networks. Deep learning architectures for graphstructured
Representation Learning on Networks, snap.stanford.edu/proj/embeddings-www, WWW 2018 1 This Talk 1) Node embeddings Map nodes to low-dimensional embeddings. 2) Graph neural networks Deep learning architectures
More informationAnomaly Detection. You Chen
Anomaly Detection You Chen 1 Two questions: (1) What is Anomaly Detection? (2) What are Anomalies? Anomaly detection refers to the problem of finding patterns in data that do not conform to expected behavior
More informationDeepWalk: Online Learning of Social Representations
DeepWalk: Online Learning of Social Representations ACM SIG-KDD August 26, 2014, Rami Al-Rfou, Steven Skiena Stony Brook University Outline Introduction: Graphs as Features Language Modeling DeepWalk Evaluation:
More informationRecommender Systems: User Experience and System Issues
Recommender Systems: User Experience and System ssues Joseph A. Konstan University of Minnesota konstan@cs.umn.edu http://www.grouplens.org Summer 2005 1 About me Professor of Computer Science & Engineering,
More informationLink Prediction for Social Network
Link Prediction for Social Network Ning Lin Computer Science and Engineering University of California, San Diego Email: nil016@eng.ucsd.edu Abstract Friendship recommendation has become an important issue
More informationIndex. Springer Nature Switzerland AG 2019 B. Moons et al., Embedded Deep Learning,
Index A Algorithmic noise tolerance (ANT), 93 94 Application specific instruction set processors (ASIPs), 115 116 Approximate computing application level, 95 circuits-levels, 93 94 DAS and DVAS, 107 110
More informationLearning to Match. Jun Xu, Zhengdong Lu, Tianqi Chen, Hang Li
Learning to Match Jun Xu, Zhengdong Lu, Tianqi Chen, Hang Li 1. Introduction The main tasks in many applications can be formalized as matching between heterogeneous objects, including search, recommendation,
More informationAspEm: Embedding Learning by Aspects in Heterogeneous Information Networks
AspEm: Embedding Learning by Aspects in Heterogeneous Information Networks Yu Shi, Huan Gui, Qi Zhu, Lance Kaplan, Jiawei Han University of Illinois at Urbana-Champaign (UIUC) Facebook Inc. U.S. Army Research
More informationData Mining Techniques
Data Mining Techniques CS 6 - Section - Spring 7 Lecture Jan-Willem van de Meent (credit: Andrew Ng, Alex Smola, Yehuda Koren, Stanford CS6) Project Project Deadlines Feb: Form teams of - people 7 Feb:
More informationVoid main Technologies
SNO TITLE Domain 1. A Hybrid Approach for Detecting Automated Spammers in Twitter Data mining 2. A Key-Policy Attribute-Based Temporary Keyword Search scheme for Secure Storage 3. A Lightweight Secure
More informationCS246: Mining Massive Datasets Jure Leskovec, Stanford University
CS6: Mining Massive Datasets Jure Leskovec, Stanford University http://cs6.stanford.edu //8 Jure Leskovec, Stanford CS6: Mining Massive Datasets High dim. data Graph data Infinite data Machine learning
More informationTowards a hybrid approach to Netflix Challenge
Towards a hybrid approach to Netflix Challenge Abhishek Gupta, Abhijeet Mohapatra, Tejaswi Tenneti March 12, 2009 1 Introduction Today Recommendation systems [3] have become indispensible because of the
More informationRecommender Systems - Content, Collaborative, Hybrid
BOBBY B. LYLE SCHOOL OF ENGINEERING Department of Engineering Management, Information and Systems EMIS 8331 Advanced Data Mining Recommender Systems - Content, Collaborative, Hybrid Scott F Eisenhart 1
More informationThanks to Jure Leskovec, Anand Rajaraman, Jeff Ullman
Thanks to Jure Leskovec, Anand Rajaraman, Jeff Ullman http://www.mmds.org Overview of Recommender Systems Content-based Systems Collaborative Filtering J. Leskovec, A. Rajaraman, J. Ullman: Mining of Massive
More informationK- Nearest Neighbors(KNN) And Predictive Accuracy
Contact: mailto: Ammar@cu.edu.eg Drammarcu@gmail.com K- Nearest Neighbors(KNN) And Predictive Accuracy Dr. Ammar Mohammed Associate Professor of Computer Science ISSR, Cairo University PhD of CS ( Uni.
More informationProbabilistic Models in Social Network Analysis
Probabilistic Models in Social Network Analysis Sargur N. Srihari University at Buffalo, The State University of New York USA US-India Workshop on Large Scale Data Analytics and Intelligent Services December
More informationDeep Model Adaptation using Domain Adversarial Training
Deep Model Adaptation using Domain Adversarial Training Victor Lempitsky, joint work with Yaroslav Ganin Skolkovo Institute of Science and Technology ( Skoltech ) Moscow region, Russia Deep supervised
More informationCS246: Mining Massive Datasets Jure Leskovec, Stanford University
CS6: Mining Massive Datasets Jure Leskovec, Stanford University http://cs6.stanford.edu Customer X Buys Metalica CD Buys Megadeth CD Customer Y Does search on Metalica Recommender system suggests Megadeth
More informationUSING OF THE K NEAREST NEIGHBOURS ALGORITHM (k-nns) IN THE DATA CLASSIFICATION
USING OF THE K NEAREST NEIGHBOURS ALGORITHM (k-nns) IN THE DATA CLASSIFICATION Gîlcă Natalia, Roșia de Amaradia Technological High School, Gorj, ROMANIA Gîlcă Gheorghe, Constantin Brîncuși University from
More informationSupervised Random Walks
Supervised Random Walks Pawan Goyal CSE, IITKGP September 8, 2014 Pawan Goyal (IIT Kharagpur) Supervised Random Walks September 8, 2014 1 / 17 Correlation Discovery by random walk Problem definition Estimate
More informationPerformance Evaluation of Various Classification Algorithms
Performance Evaluation of Various Classification Algorithms Shafali Deora Amritsar College of Engineering & Technology, Punjab Technical University -----------------------------------------------------------***----------------------------------------------------------
More informationMining of Massive Datasets Jure Leskovec, Anand Rajaraman, Jeff Ullman Stanford University Infinite data. Filtering data streams
/9/7 Note to other teachers and users of these slides: We would be delighted if you found this our material useful in giving your own lectures. Feel free to use these slides verbatim, or to modify them
More informationGraphNet: Recommendation system based on language and network structure
GraphNet: Recommendation system based on language and network structure Rex Ying Stanford University rexying@stanford.edu Yuanfang Li Stanford University yli03@stanford.edu Xin Li Stanford University xinli16@stanford.edu
More informationStudy on Recommendation Systems and their Evaluation Metrics PRESENTATION BY : KALHAN DHAR
Study on Recommendation Systems and their Evaluation Metrics PRESENTATION BY : KALHAN DHAR Agenda Recommendation Systems Motivation Research Problem Approach Results References Business Motivation What
More informationDataSToRM: Data Science and Technology Research Environment
The Future of Advanced (Secure) Computing DataSToRM: Data Science and Technology Research Environment This material is based upon work supported by the Assistant Secretary of Defense for Research and Engineering
More informationBroad Learning via Fusion of Heterogeneous Information
Broad Learning via Fusion of Heterogeneous Information Philip S. Yu Distinguished Professor & Wexler Chair University of Illinois at Chicago Dean, Institute for Data Science Tsinghua University Big Data
More informationWeighted Alternating Least Squares (WALS) for Movie Recommendations) Drew Hodun SCPD. Abstract
Weighted Alternating Least Squares (WALS) for Movie Recommendations) Drew Hodun SCPD Abstract There are two common main approaches to ML recommender systems, feedback-based systems and content-based systems.
More informationCSE 258 Lecture 8. Web Mining and Recommender Systems. Extensions of latent-factor models, (and more on the Netflix prize)
CSE 258 Lecture 8 Web Mining and Recommender Systems Extensions of latent-factor models, (and more on the Netflix prize) Summary so far Recap 1. Measuring similarity between users/items for binary prediction
More informationDay 3 Lecture 1. Unsupervised Learning
Day 3 Lecture 1 Unsupervised Learning Semi-supervised and transfer learning Myth: you can t do deep learning unless you have a million labelled examples for your problem. Reality You can learn useful representations
More informationRecommender Systems New Approaches with Netflix Dataset
Recommender Systems New Approaches with Netflix Dataset Robert Bell Yehuda Koren AT&T Labs ICDM 2007 Presented by Matt Rodriguez Outline Overview of Recommender System Approaches which are Content based
More informationELEC6910Q Analytics and Systems for Social Media and Big Data Applications Lecture 4. Prof. James She
ELEC6910Q Analytics and Systems for Social Media and Big Data Applications Lecture 4 Prof. James She james.she@ust.hk 1 Selected Works of Activity 4 2 Selected Works of Activity 4 3 Last lecture 4 Mid-term
More informationIntroduction p. 1 What is the World Wide Web? p. 1 A Brief History of the Web and the Internet p. 2 Web Data Mining p. 4 What is Data Mining? p.
Introduction p. 1 What is the World Wide Web? p. 1 A Brief History of the Web and the Internet p. 2 Web Data Mining p. 4 What is Data Mining? p. 6 What is Web Mining? p. 6 Summary of Chapters p. 8 How
More informationOn hybrid modular recommendation systems for video streaming
On hybrid modular recommendation systems for video streaming Evripides Tzamousis Maria Papadopouli arxiv:1901.01418v1 [cs.ir] 5 Jan 2019 Abstract The technological advances in networking, mobile computing,
More informationIntroduction. Chapter Background Recommender systems Collaborative based filtering
ii Abstract Recommender systems are used extensively today in many areas to help users and consumers with making decisions. Amazon recommends books based on what you have previously viewed and purchased,
More informationRecommendation Systems
Recommendation Systems CS 534: Machine Learning Slides adapted from Alex Smola, Jure Leskovec, Anand Rajaraman, Jeff Ullman, Lester Mackey, Dietmar Jannach, and Gerhard Friedrich Recommender Systems (RecSys)
More informationPerformance Comparison of Algorithms for Movie Rating Estimation
Performance Comparison of Algorithms for Movie Rating Estimation Alper Köse, Can Kanbak, Noyan Evirgen Research Laboratory of Electronics, Massachusetts Institute of Technology Department of Electrical
More informationCS246: Mining Massive Datasets Jure Leskovec, Stanford University
CS6: Mining Massive Datasets Jure Leskovec, Stanford University http://cs6.stanford.edu Training data 00 million ratings, 80,000 users, 7,770 movies 6 years of data: 000 00 Test data Last few ratings of
More informationA Recommender System Based on Improvised K- Means Clustering Algorithm
A Recommender System Based on Improvised K- Means Clustering Algorithm Shivani Sharma Department of Computer Science and Applications, Kurukshetra University, Kurukshetra Shivanigaur83@yahoo.com Abstract:
More informationBordaRank: A Ranking Aggregation Based Approach to Collaborative Filtering
BordaRank: A Ranking Aggregation Based Approach to Collaborative Filtering Yeming TANG Department of Computer Science and Technology Tsinghua University Beijing, China tym13@mails.tsinghua.edu.cn Qiuli
More informationProwess Improvement of Accuracy for Moving Rating Recommendation System
2017 IJSRST Volume 3 Issue 1 Print ISSN: 2395-6011 Online ISSN: 2395-602X Themed Section: Scienceand Technology Prowess Improvement of Accuracy for Moving Rating Recommendation System P. Damodharan *1,
More informationOnline Social Networks and Media
Online Social Networks and Media Absorbing Random Walks Link Prediction Why does the Power Method work? If a matrix R is real and symmetric, it has real eigenvalues and eigenvectors: λ, w, λ 2, w 2,, (λ
More informationSTREAMING RANKING BASED RECOMMENDER SYSTEMS
STREAMING RANKING BASED RECOMMENDER SYSTEMS Weiqing Wang, Hongzhi Yin, Zi Huang, Qinyong Wang, Xingzhong Du, Quoc Viet Hung Nguyen University of Queensland, Australia & Griffith University, Australia July
More informationarxiv: v4 [cs.ir] 28 Jul 2016
Review-Based Rating Prediction arxiv:1607.00024v4 [cs.ir] 28 Jul 2016 Tal Hadad Dept. of Information Systems Engineering, Ben-Gurion University E-mail: tah@post.bgu.ac.il Abstract Recommendation systems
More informationCombining Review Text Content and Reviewer-Item Rating Matrix to Predict Review Rating
Combining Review Text Content and Reviewer-Item Rating Matrix to Predict Review Rating Dipak J Kakade, Nilesh P Sable Department of Computer Engineering, JSPM S Imperial College of Engg. And Research,
More informationData Mining. Introduction. Hamid Beigy. Sharif University of Technology. Fall 1395
Data Mining Introduction Hamid Beigy Sharif University of Technology Fall 1395 Hamid Beigy (Sharif University of Technology) Data Mining Fall 1395 1 / 21 Table of contents 1 Introduction 2 Data mining
More information1 Overview Definitions (read this section carefully) 2
MLPerf User Guide Version 0.5 May 2nd, 2018 1 Overview 2 1.1 Definitions (read this section carefully) 2 2 General rules 3 2.1 Strive to be fair 3 2.2 System and framework must be consistent 4 2.3 System
More informationINTRODUCTION TO MACHINE LEARNING. Measuring model performance or error
INTRODUCTION TO MACHINE LEARNING Measuring model performance or error Is our model any good? Context of task Accuracy Computation time Interpretability 3 types of tasks Classification Regression Clustering
More informationarxiv: v1 [stat.ml] 13 May 2018
EXTENDABLE NEURAL MATRIX COMPLETION Duc Minh Nguyen, Evaggelia Tsiligianni, Nikos Deligiannis Vrije Universiteit Brussel, Pleinlaan 2, B-1050 Brussels, Belgium imec, Kapeldreef 75, B-3001 Leuven, Belgium
More informationA Deep Relevance Matching Model for Ad-hoc Retrieval
A Deep Relevance Matching Model for Ad-hoc Retrieval Jiafeng Guo 1, Yixing Fan 1, Qingyao Ai 2, W. Bruce Croft 2 1 CAS Key Lab of Web Data Science and Technology, Institute of Computing Technology, Chinese
More informationUse of Synthetic Data in Testing Administrative Records Systems
Use of Synthetic Data in Testing Administrative Records Systems K. Bradley Paxton and Thomas Hager ADI, LLC 200 Canal View Boulevard, Rochester, NY 14623 brad.paxton@adillc.net, tom.hager@adillc.net Executive
More informationRecommendation System for Sports Videos
Recommendation System for Sports Videos Choosing Recommendation System Approach in a Domain with Limited User Interaction Data Simen Røste Odden Master s Thesis Informatics: Programming and Networks 60
More informationAvailable online at ScienceDirect. Procedia Technology 17 (2014 )
Available online at www.sciencedirect.com ScienceDirect Procedia Technology 17 (2014 ) 528 533 Conference on Electronics, Telecommunications and Computers CETC 2013 Social Network and Device Aware Personalized
More informationMining Web Data. Lijun Zhang
Mining Web Data Lijun Zhang zlj@nju.edu.cn http://cs.nju.edu.cn/zlj Outline Introduction Web Crawling and Resource Discovery Search Engine Indexing and Query Processing Ranking Algorithms Recommender Systems
More informationAssignment 5: Collaborative Filtering
Assignment 5: Collaborative Filtering Arash Vahdat Fall 2015 Readings You are highly recommended to check the following readings before/while doing this assignment: Slope One Algorithm: https://en.wikipedia.org/wiki/slope_one.
More informationIntroduction to Data Mining
Introduction to Data Mining Lecture #7: Recommendation Content based & Collaborative Filtering Seoul National University In This Lecture Understand the motivation and the problem of recommendation Compare
More informationA PROPOSED HYBRID BOOK RECOMMENDER SYSTEM
A PROPOSED HYBRID BOOK RECOMMENDER SYSTEM SUHAS PATIL [M.Tech Scholar, Department Of Computer Science &Engineering, RKDF IST, Bhopal, RGPV University, India] Dr.Varsha Namdeo [Assistant Professor, Department
More informationData Mining. Introduction. Hamid Beigy. Sharif University of Technology. Fall 1394
Data Mining Introduction Hamid Beigy Sharif University of Technology Fall 1394 Hamid Beigy (Sharif University of Technology) Data Mining Fall 1394 1 / 20 Table of contents 1 Introduction 2 Data mining
More informationLecture 11: Clustering Introduction and Projects Machine Learning
Lecture 11: Clustering Introduction and Projects Machine Learning Andrew Rosenberg March 12, 2010 1/1 Last Time Junction Tree Algorithm Efficient Marginals in Graphical Models 2/1 Today Clustering Project
More informationComparison of Recommender System Algorithms focusing on the New-Item and User-Bias Problem
Comparison of Recommender System Algorithms focusing on the New-Item and User-Bias Problem Stefan Hauger 1, Karen H. L. Tso 2, and Lars Schmidt-Thieme 2 1 Department of Computer Science, University of
More informationCS224W: Analysis of Networks Jure Leskovec, Stanford University
CS224W: Analysis of Networks Jure Leskovec, Stanford University http://cs224w.stanford.edu Jure Leskovec, Stanford CS224W: Analysis of Networks, http://cs224w.stanford.edu 2????? Machine Learning Node
More informationMachine Learning - Clustering. CS102 Fall 2017
Machine Learning - Fall 2017 Big Data Tools and Techniques Basic Data Manipulation and Analysis Performing well-defined computations or asking well-defined questions ( queries ) Data Mining Looking for
More informationAutomatically Building Research Reading Lists
Automatically Building Research Reading Lists Michael D. Ekstrand 1 Praveen Kanaan 1 James A. Stemper 2 John T. Butler 2 Joseph A. Konstan 1 John T. Riedl 1 ekstrand@cs.umn.edu 1 GroupLens Research Department
More informationWeb-Scale Image Search and Their Applications
Web-Scale Image Search and Their Applications Sung-Eui Yoon KAIST http://sglab.kaist.ac.kr Project Guidelines: Project Topics Any topics related to the course theme are okay You can find topics by browsing
More informationRecommender Systems. Master in Computer Engineering Sapienza University of Rome. Carlos Castillo
Recommender Systems Class Program University Semester Slides by Data Mining Master in Computer Engineering Sapienza University of Rome Fall 07 Carlos Castillo http://chato.cl/ Sources: Ricci, Rokach and
More informationRapid growth of massive datasets
Overview Rapid growth of massive datasets E.g., Online activity, Science, Sensor networks Data Distributed Clusters are Pervasive Data Distributed Computing Mature Methods for Common Problems e.g., classification,
More informationG(B)enchmark GraphBench: Towards a Universal Graph Benchmark. Khaled Ammar M. Tamer Özsu
G(B)enchmark GraphBench: Towards a Universal Graph Benchmark Khaled Ammar M. Tamer Özsu Bioinformatics Software Engineering Social Network Gene Co-expression Protein Structure Program Flow Big Graphs o
More informationInternational Journal of Scientific Research & Engineering Trends Volume 4, Issue 6, Nov-Dec-2018, ISSN (Online): X
Analysis about Classification Techniques on Categorical Data in Data Mining Assistant Professor P. Meena Department of Computer Science Adhiyaman Arts and Science College for Women Uthangarai, Krishnagiri,
More informationMachine Learning in the Process Industry. Anders Hedlund Analytics Specialist
Machine Learning in the Process Industry Anders Hedlund Analytics Specialist anders@binordic.com Artificial Specific Intelligence Artificial General Intelligence Strong AI Consciousness MEDIA, NEWS, CELEBRITIES
More informationCS 124/LINGUIST 180 From Languages to Information
CS /LINGUIST 80 From Languages to Information Dan Jurafsky Stanford University Recommender Systems & Collaborative Filtering Slides adapted from Jure Leskovec Recommender Systems Customer X Buys Metallica
More informationCS 124/LINGUIST 180 From Languages to Information
CS /LINGUIST 80 From Languages to Information Dan Jurafsky Stanford University Recommender Systems & Collaborative Filtering Slides adapted from Jure Leskovec Recommender Systems Customer X Buys CD of
More information