Learning optimal ranking with tensor factorization for tag recommendation
By Steffen Rendle, Leandro Bezerra Marinho, Alexandros Nanopoulos, Lars Schmidt-Thieme. Knowledge Discovery and Data Mining, 2009
398 citations recorded by the source corpus.
Subfields: Computational Mathematics
View Paper: https://www.semanticscholar.org/paper/24b06b15cc18b4b44738747380108ba72bbc3ada
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computational Mathematics
- Steffen Rendle
- Sophie Ahrens
- Yue Shi
- Martha A. Larson
- Alan Hanjalic
- Lars Schmidt-Thieme
- Yongfeng Zhang
- Xu Chen
Citations in the corpus
- Recommender Systems (2012, methodology): “They implement tensor factorization [280] to reduce the ternary relation into low-rank feature matrices, alleviate the sparsity problem in large-scale datasets and ultimately provide personalized recommendations [281, 282].”
- Pairwise interaction tensor factorization for personalized tag recommendation (2010, methodology): “In [18] a better learning approach for TD models has been introduced, that optimizes the model parameters for the ranking statistic AUC (area under the ROC-curve).”
- 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.”
- Collaborative Filtering beyond the User-Item Matrix (2014, methodology): “These authors later formalized this ranking criterion with their Bayesian personalized ranking (BPR) [Rendle et al. 2009b] approach.”
- Personalized Deep Learning for Tag Recommendation (2017, neutral): “Let PS := {(u, i) | ∃t ∈ T : (u, i, t) ∈ S} be all observed posts [17].”
- Recommender Systems (2016, neutral): “Tensor-based methods for tag recommendation are presented in [497, 498, 582, 583].”
- Compositional Matrix-Space Models of Language (2010, methodology): “Then tensor decomposition techniques can be applied in order to find a compact representation, reduce noise, and cluster together similar tokens (Tucker, 1966; Rendle et al., 2009).”