列出了librec跑出的rating-prediction的结果 (TrustSVD: Collaborative Filtering with Both the Explicit and Implicit Influence of User Trust and of Item Ratings)
| algorithm | short name | directory path | superClass |
|---|---|---|---|
| ConstantGuessRecommender | constantguess | baseline | AbstractRecommender |
| GlobalAverageRecommender | globalaverage | baseline | AbstractRecommender |
| ItemAverageRecommender | itemaverage | baseline | AbstractRecommender |
| ItemClusterRecommender | itemcluster | baseline | ProbabilisticGraphicalRecommender |
| MostPopularRecommender | mostpopular | baseline | AbstractRecommender |
| RandomGuessRecommender | randomguess | baseline | AbstractRecommender |
| UserAverageRecommender | useraverage | baseline | AbstractRecommender |
| UserClusterRecommender | usercluster | baseline | ProbabilisticGraphicalRecommender |
| AoBPRRecommender | aobpr | cf.ranking | MatrixFactorizationRecommender |
| Rendle S, Freudenthaler C. Improving pairwise learning for item recommendation from implicit feedback[C]// WSDM 2014 | |||
| AspectModelRecommender | aspectmodelranking | cf.ranking | ProbabilisticGraphicalRecommender |
| Hofmann T, Puzicha J. Latent class models for collaborative filtering[C]//IJCAI. 1999 | |||
| BPRRecommender | bpr | cf.ranking | MatrixFactorizationRecommender |
| Rendle S, Freudenthaler C, Gantner Z, et al. BPR: Bayesian personalized ranking from implicit feedback[C]// AUAI 2009 | |||
| CLIMFRecommender | climf | cf.ranking | MatrixFactorizationRecommender |
| Shi Y, Karatzoglou A, Baltrunas L, et al. CLiMF: learning to maximize reciprocal rank with collaborative less-is-more filtering[C]//RecSys. 2012 | |||
| EALSRecommender | eals | cf.ranking | MatrixFactorizationRecommender |
| He X, Zhang H, Kan M Y, et al. Fast matrix factorization for online recommendation with implicit feedback[C]//SIGIR. 2016 | |||
| FISMaucRecommender | fismauc | cf.ranking | MatrixFactorizationRecommender |
| Kabbur S, Ning X, Karypis G. Fism: factored item similarity models for top-n recommender systems[C]//KDD. 2013 | |||
| FISMrmseRecommender | fismrmse | cf.ranking | MatrixFactorizationRecommender |
| Kabbur S, Ning X, Karypis G. Fism: factored item similarity models for top-n recommender systems[C]//KDD. 2013 | |||
| GBPRRecommender | gbpr | cf.ranking | MatrixFactorizationRecommender |
| Pan W, Chen L . GBPR: Group Preference Based Bayesian Personalized Ranking for One-Class Collaborative Filtering[C]//IJCAI. 2013 | |||
| ItemBigramRecommender | itembigram | cf.ranking | ProbabilisticGraphicalRecommender |
| Wallach H M. Topic modeling: beyond bag-of-words[C]//ICML2016 | |||
| LDARecommender | lda | cf.ranking | ProbabilisticGraphicalRecommender |
| Griffiths T. Gibbs sampling in the generative model of latent dirichlet allocation[J]// TOIS 2014 | |||
| ListwiseMFRecommender | Listwisemf | cf.ranking | MatrixFactorizationRecommender |
| Shi Y, Larson M, Hanjalic A. List-wise learning to rank with matrix factorization for collaborative filtering[C]//RecSys 2010 | |||
| PLSARecommender | plsa | cf.ranking | ProbabilisticGraphicalRecommender |
| Hofmann T. Latent semantic models for collaborative filtering[J]// TOIS. 2014 | |||
| RankALSRecommender | rankals | cf.ranking | MatrixFactorizationRecommender |
| Takács G, Tikk D. Alternating least squares for personalized ranking[C]//RecSys. 2012 | |||
| RankSGDRecommender | ranksgd | cf.ranking | MatrixFactorizationRecommender |
| M Jahrer. Collaborative Filtering Ensemble for Ranking.// JMLR 2012 | |||
| SLIMRecommender | slim | cf.ranking | AbstractRecommender |
| Ning X, Karypis G. Slim: Sparse linear methods for top-n recommender systems[C]//ICDM 2011 | |||
| WBPRRecommender | wbpr | cf.ranking | MatrixFactorizationRecommender |
| Gantner Z, Drumond L, Freudenthaler C, et al. Personalized Ranking for Non-Uniformly Sampled Items[C]//KDD Cup. 2012 | |||
| WRMFRecommender | wrmf | cf.ranking | MatrixFactorizationRecommender |
| Hu Y, Koren Y, Volinsky C. Collaborative filtering for implicit feedback datasets[C]//ICDM. 2008 | |||
| AspectModelRecommender | aspectmodelrating | cf.rating | ProbabilisticGraphicalRecommender |
| Hofmann T, Puzicha J. Latent class models for collaborative filtering[C]//IJCAI. 1999 | |||
| ASVDPlusPlusRecommender | asvdpp | cf.rating | BiasedMFRecommender → MatrixFactorizationRecommender |
| Koren Y. Factorization meets the neighborhood: a multifaceted collaborative filtering model[C]//KDD. 2008 | |||
| BiasedMFRecommender | biasedmf | cf.rating | MatrixFactorizationRecommender |
| Koren Y, Bell R, Volinsky C. Matrix factorization techniques for recommender systems[J]//Computer, 2009 | |||
| BNPoissMFRecommender | bnpoissmf | cf.rating | MatrixFactorizationRecommender |
| BPMFRecommender | bpmf | cf.rating | MatrixFactorizationRecommender |
| Salakhutdinov R, Mnih A. Bayesian probabilistic matrix factorization using Markov chain Monte Carlo[C]//ICML 2008 | |||
| BPoissMFRecommender | bpoissmf | cf.rating | MatrixFactorizationRecommender |
| Gopalan P, Hofman J M, Blei D M. Scalable recommendation with poisson factorization[J]// arXiv 2013. | |||
| FMALSRecommender | fmals | cf.rating | FactorizationMachineRecommender |
| Rendle S. Factorization machines with libfm[J]//TITS 2012 | |||
| FMSGDRecommender | fmsgd | cf.rating | FactorizationMachineRecommender |
| Rendle S. Factorization machines with libfm[J]//TITS 2012 | |||
| GPLSARecommender | gplsa | cf.rating | ProbabilisticGraphicalRecommender |
| Hofmann T. Collaborative filtering via gaussian probabilistic latent semantic analysis[C]//SIGIR 2003 | |||
| LDCCRecommender | ldcc | cf.rating | ProbabilisticGraphicalRecommender |
| Wang P, Domeniconi C, Laskey K B. Latent dirichlet bayesian co-clustering[C]// Database 2009 | |||
| LLORMARecommender | llorma | cf.rating | MatrixFactorizationRecommender |
| Lee J, Kim S, Lebanon G, et al. Local Low-Rank Matrix Approximation[J]// ICML 2013 | |||
| MFALSRecommender | mfals | cf.rating | MatrixFactorizationRecommender |
| Zhou Y, Wilkinson D, Schreiber R, et al. Large-scale parallel collaborative filtering for the netflix prize[C]//AAIM 2008 | |||
| NMFRecommender | nmf | cf.rating | MatrixFactorizationRecommender |
| Lee D D, Seung H S. Algorithms for non-negative matrix factorization[C]//NIPS 2001 | |||
| PMFRecommender | pmf | cf.rating | MatrixFactorizationRecommender |
| Salakhutdinov R, Mnih A. Probabilistic Matrix Factorization[C]//NIPS 2007 | |||
| RBMRecommender | rbm | cf.rating | MatrixFactorizationRecommender |
| Salakhutdinov R, Mnih A, Hinton G. Restricted Boltzmann machines for collaborative filtering[C]//ICML2007 | |||
| RFRecRecommender | rfrec | cf.rating | MatrixFactorizationRecommender |
| Gedikli F, Bagdat F, Ge M, et al. RF-REC: Fast and accurate computation of recommendations based on rating frequencies[C]//IEEE(CEC) 2011 | |||
| SVDPlusPlusRecommender | svdpp | cf.rating | BiasedMFRecommender → MatrixFactorizationRecommender |
| Koren Y. Factorization meets the neighborhood: a multifaceted collaborative filtering model[C]//KDD 2008 | |||
| URPRecommender | urp | cf.rating | ProbabilisticGraphicalRecommender |
| Marlin B M. Modeling User Rating Profiles For Collaborative Filtering[C]//NIPS. 2003 | |||
| BHFreeRecommender | bhfree | cf | ProbabilisticGraphicalRecommender |
| Barbieri N, Manco G, Ortale R, et al. Balancing prediction and recommendation accuracy: hierarchical latent factors for preference data[C]//SDM 2012 | |||
| BUCMRecommender | bucm | cf | ProbabilisticGraphicalRecommender |
| Barbieri N, Costa G, Manco G, et al. Modeling item selection and relevance for accurate recommendations: a bayesian approach[C]//RecSys 2011 | |||
| ItemKNNRecommender | itemknn | cf | AbstractRecommender |
| Deshpande M, Karypis G. Item-based top-n recommendation algorithms[J]//TOIS 2004 | |||
| UserKNNRecommender | userknn | cf | AbstractRecommender |
| Konstan J A, Miller B N, Maltz D, et al. GroupLens: applying collaborative filtering to Usenet news[J]//Communications of the ACM, 1997 | |||
| EFMRecommender | efm | content | BiasedMFRecommender → MatrixFactorizationRecommender |
| Zhang Y, Lai G, Zhang M, et al. Explicit factor models for explainable recommendation based on phrase-level sentiment analysis[C]//SIGIR 2014 | |||
| HFTRecommender | hft | content | BiasedMFRecommender → MatrixFactorizationRecommender |
| McAuley J, Leskovec J. Hidden factors and hidden topics: understanding rating dimensions with review text[C]//RecSys 2013 | |||
| SBPRRecommender | sbpr | context.ranking | SocialRecommender |
| Zhao T, McAuley J, King I. Leveraging social connections to improve personalized ranking for collaborative filtering[C]//CIKM 2014 | |||
| BPTFRecommender | bptf | context.rating | TensorRecommender |
| Schein A, Paisley J, Blei D M, et al. Bayesian poisson tensor factorization for inferring multilateral relations from sparse dyadic event counts[C]//SIGKDD 2015 | |||
| PITFRecommender | pitf | context.rating | TensorRecommender |
| Rendle S, Schmidt-Thieme L. Pairwise interaction tensor factorization for personalized tag recommendation[C]//WSDM 2010 | |||
| RSTERecommender | rste | context.rating | SocialRecommender |
| Ma H, King I, Lyu M R. Learning to recommend with social trust ensemble[C]//SIGIR 2009 | |||
| SocialMFRecommender | socialmf | context.rating | SocialRecommender |
| Jamali M, Ester M. A matrix factorization technique with trust propagation for recommendation in social networks[C]//RecSys 2010 | |||
| SoRecRecommender | sorec | context.rating | SocialRecommender |
| Jamali M, Ester M. A matrix factorization technique with trust propagation for recommendation in social networks[C]//RecSys 2010 | |||
| SoRegRecommender | soreg | context.rating | SocialRecommender |
| Ma H, Zhou D, Liu C, et al. Recommender systems with social regularization[C]//WSDM 2011 | |||
| TimeSVDRecommender | timesvd | context.rating | BiasedMFRecommender → MatrixFactorizationRecommender |
| Koren Y. Collaborative filtering with temporal dynamics[J]. Communications of the ACM, 2010, 53(4): 89-97. | |||
| TrustMFRecommender | trustmf | context.rating | SocialMFRecommender |
| Yang B, Lei Y, Liu J, et al. Social collaborative filtering by trust[J]//IJCAI 2016. | |||
| TrustSVDRecommender | trustsvd | context.rating | SocialRecommender |
| Guo G, Zhang J, Yorke-Smith N. TrustSVD: Collaborative Filtering with Both the Explicit and Implicit Influence of User Trust and of Item Ratings[C]//Aaai 2015 | |||
| AssociationRuleRecommender | associationrule | ext | AbstractRecommender |
| Kim C, Kim J. A recommendation algorithm using multi-level association rules[C]//WI 2003 | |||
| ExternalRecommender | external | ext | AbstractRecommender |
| PersonalityDiagnosisRecommender | personalitydiagnosis | ext | AbstractRecommender |
| PRankDRecommender | prankd | ext | RankSGDRecommender → MatrixFactorizationRecommender |
| Hurley N J. Personalised ranking with diversity[C]//RecSys 2013 | |||
| SlopeOneRecommender | slopeone | ext | AbstractRecommender |
| Lemire D, Maclachlan A. Slope one predictors for online rating-based collaborative filtering[C]//SIAM 2005 | |||
| HybridRecommender | hybrid | hybrid | AbstractRecommender |
| Zhou T, Kuscsik Z, Liu J G, et al. Solving the apparent diversity-accuracy dilemma of recommender systems[J]//PNAS. 2010 | |||
### Algorithm Configuration List #### Baseline ##### ConstantGuessRecommender ``` rec.recommender.class=constantguess ``` ##### GlobalAverageRecommender ``` rec.recommender.class=globalaverage ``` ##### ItemAverageRecommender ``` rec.recommender.class=itemaverage ``` ##### ItemClusterRecommender ``` rec.recommender.class=itemcluster rec.pgm.number=10 rec.iterator.maximum=20 ``` ##### MostPopularRecommender ``` rec.recommender.class=mostpopular ``` ##### RandomGuessRecommender ``` rec.recommender.class=randomguess ``` ##### UserAverageRecommender ``` rec.recommender.class=useraverage ``` ##### UserClusterRecommender ``` rec.recommender.class=usercluster rec.factory.number=10 rec.iterator.maximum=20 ``` #### Collaborative Filtering (item ranking) ##### AOBPRRecommender ``` rec.recommender.class=aobpr rec.item.distribution.parameter = 500 rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=100 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 rec.recommender.isranking=true rec.recommender.ranking.topn=10 ``` ##### AspectModelRecommender ``` rec.recommender.class=aspectmodelranking rec.iterator.maximum=20 rec.recommender.isranking=true rec.recommender.ranking.topn=10 data.splitter.cv.number=5 rec.pgm.burnin=10 rec.pgm.samplelag=10 rec.topic.number=10 ``` ##### BPRRecommender ``` rec.recommender.class=bpr rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=100 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.factor.number=10 rec.learnRate.bolddriver=false rec.learnRate.decay=1.0 rec.recommender.isranking=true rec.recommender.ranking.topn=10 ``` ##### CLIMFRecommender ``` rec.recommender.class=climf rec.iterator.learnrate=0.001 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=100 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 rec.recommender.isranking=true rec.recommender.ranking.topn=10 ``` ##### EALSRecommender ``` rec.recommender.class=eals rec.iterator.maximum=10 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.factor.number=10 rec.recommender.isranking=true rec.recommender.ranking.topn=10 # 0：eALS MF; 1：WRMF; 2: both rec.eals.wrmf.judge=1 # the overall weight of missing data c0 rec.eals.overall=128 # the significance level of popular items over un-popular ones rec.eals.ratio=0.4 # confidence weight coefficient, alpha in original paper rec.wrmf.weight.coefficient=4.0 ``` ##### FISMaucRecommender ``` rec.recommender.class=fismauc rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=10 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 rec.recommender.isranking=true rec.recommender.ranking.topn=10 rec.fismauc.rho=2 rec.fismauc.alpha=1.5 ``` ##### FISMrmseRecommender ``` rec.recommender.class=fismrmse rec.iteration.learnrate=0.01 rec.iterator.maximum=100 rec.recommender.isranking=true rec.fismrmse.rho=1 rec.fismrmse.alpha=1.5 ``` ##### GBPRRecommender ``` rec.recommender.class=gbpr rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=100 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 rec.recommender.isranking=true rec.recommender.ranking.topn=10 ``` ##### ItemBigramRecommender ``` rec.recommender.class=itembigram data.column.format=UIRT data.input.path=test/ratings-date.txt rec.iterator.maximum=100 rec.topic.number=10 rec.recommender.isranking=true rec.recommender.ranking.topn=10 rec.user.dirichlet.prior=0.01 rec.topic.dirichlet.prior=0.01 rec.pgm.burnin=10 rec.pgm.samplelag=10 ``` ##### LDARecommender ``` rec.recommender.class=lda rec.iterator.maximum=100 rec.topic.number = 10 rec.recommender.isranking=true rec.recommender.ranking.topn=10 rec.user.dirichlet.prior=0.01 rec.topic.dirichlet.prior=0.01 rec.pgm.burnin=10 rec.pgm.samplelag=10 data.splitter.cv.number=5 # (0.0 may be a better choose than -1.0) data.convert.binarize.threshold=0.0 ``` ##### ListwiseMFRecommender ``` rec.recommender.class=listwisemf rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=100 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 rec.recommender.isranking=true rec.recommender.ranking.topn=10 ``` ##### PLSARecommender ``` rec.recommender.class=plsa rec.iteration.learnrate=0.01 rec.iterator.maximum=100 rec.recommender.isranking=true rec.topic.number = 10 rec.recommender.ranking.topn=10 # (0.0 may be a better choose than -1.0) data.convert.binarize.threshold=0.0 ``` ##### RankALSRecommender ``` rec.recommender.class=rankals rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=100 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 rec.recommender.isranking=true rec.recommender.ranking.topn=10 rec.rankals.support.weight=true ``` ##### RankSGDRecommender ``` rec.recommender.class=ranksgd rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=30 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 rec.recommender.isranking=true rec.recommender.ranking.topn=10 ``` ##### SLIMRecommender ``` rec.recommender.class=slim rec.similarity.class=cos # can only use item similarity rec.recommender.similarities=item rec.iterator.maximum=40 rec.similarity.shrinkage=10 rec.recommender.isranking=true rec.recommender.ranking.topn=10 rec.neighbors.knn.number=50 rec.recommender.earlystop=true rec.slim.regularization.l1=1 rec.slim.regularization.l2=5 ``` ##### WBPRRecommender ``` rec.recommender.class=wbpr rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=10 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.bias.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 rec.recommender.isranking=true rec.recommender.ranking.topn=10 ``` ##### WRMFRecommender ``` rec.recommender.class=wrmf rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=20 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 rec.recommender.isranking=true rec.recommender.ranking.topn=10 # confidence weight coefficient, alpha in original paper rec.wrmf.weight.coefficient=4.0 ``` #### Collaborative Filtering (rating prediction) ##### AspectModelRecommender ``` rec.recommender.class=aspectmodelrating rec.iteration.learnrate=0.01 rec.iterator.maximum=100 ``` ##### ASVDPlusPlusRecommender ``` rec.recommender.class=asvdpp rec.iteration.learnrate=0.01 rec.iterator.maximum=20 ``` ##### BiasedMFRecommender ``` rec.recommender.class=biasedmf rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=1 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.bias.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 ``` ##### BNPoissMFRecommender ``` rec.recommender.class=bnpoissmf rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=100 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 ``` ##### BPMFRecommender ``` rec.recommender.class=bpmf rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=100 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 ``` ##### BPoissMFRecommender ``` rec.recommender.class=bpoissmf rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=100 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 ``` ##### FMALSRecommender ``` data.input.path=arfftest/data.arff data.column.format=UIR data.model.splitter=ratio data.convertor.format=arff data.model.format=arff rec.recommender.class=fmals rec.iterator.learnRate=0.01 rec.iterator.maximum=100 rec.factor.number=10 ``` ##### FMSGDRecommender ``` data.input.path=arfftest/data.arff data.column.format=UIR data.model.splitter=ratio data.convertor.format=arff data.model.format=arff rec.recommender.class=fmsgd rec.iterator.learnRate=0.001 rec.iterator.maximum=100 rec.factor.number=10 ``` ##### GPLSARecommender ``` rec.recommender.class=gplsa rec.iteration.learnrate=0.01 rec.iterator.maximum=100 rec.recommender.smoothWeight=2 rec.recommender.isranking=false rec.topic.number = 10 ``` ##### LDCCRecommender ``` rec.recommender.class=ldcc rec.iteration.learnrate=0.01 rec.iterator.maximum=100 ``` ##### LLORMARecommender ``` rec.recommender.class=llorma rec.llorma.global.factors.num = 10 rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=100 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 ``` ##### MFALSRecommender ``` rec.recommender.class=mfals rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=100 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 ``` ##### NMFRecommender ``` rec.recommender.class=nmf rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=100 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 ``` ##### PMFRecommender ``` rec.recommender.class=pmf rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=50 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 ``` ##### RBMRecommender ``` rec.recommender.class=rbm rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=100 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 ``` ##### RFRecRecommender ``` rec.recommender.class=rfrec rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=100 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 ``` ##### SVDPlusPlusRecommender ``` rec.recommender.class=svdpp rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=13 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.impItem.regularization=0.001 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 ``` ##### URPRecommender ``` rec.recommender.class=urp rec.iteration.learnrate=0.01 rec.iterator.maximum=100 ``` #### Collaborative Filtering (rating prediction and item ranking) ##### BHFreeRecommender ``` rec.recommender.class=bhfree rec.pgm.burnin=10 rec.pgm.samplelag=10 rec.iterator.maximum=100 # true for item ranking, false for rating prediction rec.recommender.isranking=true rec.recommender.ranking.topn=10 ``` ##### BUCMRecommender ``` rec.recommender.class=bucm rec.pgm.burnin=10 rec.pgm.samplelag=10 rec.iterator.maximum=100 rec.pgm.topic.number=10 rec.bucm.alpha=0.01 rec.bucm.beta=0.01 rec.bucm.gamma=0.01 # true for item ranking, false for rating prediction rec.recommender.isranking=true rec.recommender.ranking.topn=10 ``` ##### ItemKNNRecommender ``` rec.recommender.class=itemknn # true for item ranking, false for rating prediction rec.recommender.isranking=false rec.recommender.ranking.topn=10 rec.recommender.similarities=item rec.similarity.class=pcc rec.neighbors.knn.number=50 rec.similarity.shrinkage=10 ``` ##### UserKNNRecommender ``` rec.similarity.class=pcc rec.neighbors.knn.number=50 rec.recommender.class=userknn rec.recommender.similarities=user # true for item ranking, false for rating prediction rec.recommender.isranking=false rec.recommender.ranking.topn=10 rec.filter.class=generic rec.similarity.shrinkage=10 ``` #### Content ##### EFMRecommender ``` data.input.path=efmtest/efm.txt rec.recommender.class=efm rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=100 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 rec.recommender.lambda.user=0.05 rec.recommender.lambda.item=0.05 rec.bias.regularization = 0.01 ``` ##### HFTRecommender ``` data.input.path=hfttest/hft.txt/ rec.recommender.class=hft rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=2 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 rec.eval.enable = 1 rec.recommender.lambda.user=0.05 rec.recommender.lambda.item=0.05 rec.bias.regularization = 0.01 ``` #### Context(item ranking) ##### SBPRRecommender ``` data.appender.class=social data.appender.path=test/test-append-dir rec.recommender.class=sbpr rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=200 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.social.regularization=0.01 rec.bias.regularization=0.01 rec.factor.number=128 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 rec.recommender.earlystop=false rec.recommender.verbose=true rec.recommender.isranking=true rec.recommender.ranking.topn=10 ``` #### Context(rating prediction) ##### BPTFRecommender ``` rec.recommender.class=bptf rec.iteration.learnrate=0.01 rec.iterator.maximum=100 ``` ##### PITFRecommender ``` rec.recommender.class=pitf rec.iteration.learnrate=0.01 rec.iterator.maximum=100 ``` ##### RSTERecommender ``` data.appender.class=social data.appender.path=test/test-append-dir rec.recommender.class=rste rec.iterator.learnrate=0.05 rec.iterator.learnrate.maximum=0.05 rec.iterator.maximum=200 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.social.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 rec.recommender.earlystop=false rec.recommender.verbose=true rec.user.social.ratio=0.8 ``` ##### SocialMFRecommender ``` data.appender.class=social data.appender.path=test/test-append-dir rec.recommender.class=socialmf rec.iterator.learnrate=0.05 rec.iterator.learnrate.maximum=0.05 rec.iterator.maximum=100 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.social.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 rec.recommender.earlystop=false rec.recommender.verbose=true ``` ##### SoRecRecommender ``` data.appender.class=social data.appender.path=test/test-append-dir rec.recommender.class=sorec rec.iterator.learnrate=0.05 rec.iterator.learnrate.maximum=0.05 rec.iterator.maximum=1000 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.rate.social.regularization=0.01 rec.user.social.regularization=0.01 rec.social.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 rec.recommender.earlystop=false rec.recommender.verbose=true ``` ##### SoRegRecommender ``` data.appender.class=social data.appender.path=test/test-append-dir rec.recommender.class=soreg rec.recommender.similarities=social rec.similarity.class=pcc rec.iterator.learnrate=0.001 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=10 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.social.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 rec.recommender.earlystop=false rec.recommender.verbose=true rec.similarity.shrinkage=10 ``` ##### TimeSVDRecommender ``` rec.recommender.class=timesvd data.column.format=UIRT data.input.path=test/ratings-date.txt rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=100 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.learnrate.decay=1.0 ``` ##### TrustMFRecommender ``` data.appender.class=social data.appender.path=test/test-append-dir rec.recommender.class=trustmf rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=30 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.social.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 rec.recommender.earlystop=false rec.recommender.verbose=true rec.social.model=T ``` ##### TrustSVDRecommender ``` data.appender.class=social data.appender.path=test/test-append-dir rec.recommender.class=trustsvd rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=50 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.social.regularization=0.01 rec.bias.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 rec.recommender.earlystop=false rec.recommender.verbose=true ``` #### Extra ##### AssociationRuleRecommender ``` rec.recommender.class=associationrule ``` ##### ExternalRecommender ``` rec.recommender.class=external ``` ##### PersonalityDiagnosisRecommender ``` rec.recommender.class=personalitydiagnosis rec.PersonalityDiagnosis.sigma=0.1 ``` ##### PRankDRecommender ``` rec.recommender.class=prankd rec.similarity.class=cos rec.recommender.similarities=item rec.similarity.shrinkage=10 rec.iterator.learnrate=0.01 rec.iterator.learnrate.maximum=0.01 rec.iterator.maximum=50 rec.user.regularization=0.01 rec.item.regularization=0.01 rec.factor.number=10 rec.learnrate.bolddriver=false rec.learnrate.decay=1.0 rec.sim.filter=4.0 ``` ##### SlopeOneRecommender ``` rec.recommender.class=slopeone rec.eval.enable=true rec.iterator.maximum=50 rec.factory.number=30 rec.iterator.learn.rate=0.001 rec.recommender.lambda.user=0.05 rec.recommender.lambda.item=0.05 ``` #### Hybrid ##### HybridRecommender ``` rec.recommender.class=hybrid rec.hybrid.lambda=0.1 rec.iterator.maximum=50 rec.factory.number=30 rec.iterator.learn.rate=0.001 rec.recommender.lambda.user=0.05 rec.recommender.lambda.item=0.05 ```
