Pairwise interaction tensor factorization for personalized tag recommendation

By Steffen Rendle, Lars Schmidt-Thieme. 2010

733 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computational Mathematics, Information Systems

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computational Mathematics

In Information Systems

Citations in the corpus

DeepFM: A Factorization-Machine based Neural Network for CTR Prediction (2017, methodology): “Besides generalized linear models and FM, a few other models are proposed for CTR prediction, such as tree-based model [ He et al. , 2014 ] , tensor based model [ Rendle and Schmidt-Thieme, 2010 ] , support vector machine [ Chang et al. , 2010 ] , and bayesian model [ Graepel et al. , 2010 ] .”
Fast context-aware recommendations with factorization machines (2011, background): “This matches to results from the related area of tag recommendation, where a pairwise interaction model (PITF) outperforms the Tucker decomposition empirically in sparse problems [15].”
Translation-based Recommendation (2017, background): “, [18, 19, 21]) make a weaker assumption that users simply prefer observed feedback over unobserved feedback and optimize the pairwise rankings of (positive, non-positive) pairs.”
Leveraging Social Connections to Improve Personalized Ranking for Collaborative Filtering (2014, neutral): “Copyright 2014 ACM 978-1-4503-2598-1/14/11 ...$15.00. http://dx.doi.org/10.1145/2661829.2661998 .”
Deriving Validity Time in Knowledge Graph (2018, methodology): “In addition, FMs can mimic many different matrix factorization models such as biased matrix factorization, Singular Value Decomposition (SVD++) [8], and Pairwise Interaction Tensor Factorization (PITF) [13].”
xDeepFM: Combining Explicit and Implicit Feature Interactions for Recommender Systems (2018, background): “Thus researchers extend the factorization models to a Bayesian Personalized Ranking (BPR) framework [10, 32, 33, 42] for implicit feedback.”
Content Driven User Profiling for Comment-Worthy Recommendations of News and Blog Articles (2015, methodology): “We take recourse to Bayesian Personalized Ranking (BPR) [9, 21] which provides a Bayesian model for learning a personalized total ranking for each user.”
Recommender Systems (2016, positive): “Tensor-based methods for tag recommendation are presented in [497, 498, 582, 583].”
Learning to Rank Features for Recommendation over Multiple Categories (2016, methodology): “A number of techniques exist for tensor factorization [15, 16, 13], and we adopt a method similar to [15] due to its eﬃciency and relatively lower learning complexity.”
Factorization Machines with libFM (2012, methodology): “…models are an extension for relations over several categorical variables; among the proposed tensor factorization approaches are Tucker Decomposition [Tucker 1966], Parallel Factor Analysis [Harshman 1970], or Pairwise Interaction Tensor factorization [Rendle and Schmidt-Thieme 2010].”
Explainable Outfit Recommendation with Joint Outfit Matching and Comment Generation (2018, neutral): “All are widely used evaluation metrics in recommender systems [56, 57, 58].”
Field-weighted Factorization Machines for Click-Through Rate Prediction in Display Advertising (2018, methodology): “Many models have been proposed to resolve this problem such as Logistic Regression (LR) [4, 16, 21], Polynomial-2 (Poly2) [3], treebased models [11], tensor-based models [20], Bayesian models [8], and Field-aware Factorization Machines (FFMs) [12, 13].”
Collaborative Filtering beyond the User-Item Matrix (2014, background): “In addition, ACM Computing Surveys, Vol. 47, No. 1, Article 3, Publication date: April 2014. the authors further proposed a pairwise interaction tensor factorization (PITF) model [Rendle and Schmidt-Thieme 2010] to improve the performance of tag recommendation.”