Scalable Tensor Decompositions for Multi-aspect Data Mining

By Tamara G. Kolda, Jimeng Sun. 2008

389 citations recorded by the source corpus.

Subfields: Computational Mathematics, Hardware and Architecture

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

In Hardware and Architecture

Citations in the corpus

Multi-Relational Latent Semantic Analysis (2013, methodology): “com We apply a memory-efficient Tucker decomposition algorithm (Kolda and Sun, 2008) implemented in tensor toolbox v2.”
Distributional Memory: A General Framework for Corpus-Based Semantics (2010, neutral): “At least as a trend, tensor decomposition appears to be better than matrix decomposition, but only marginally so (Turney does not perform this comparison).”
The structure and dynamics of multilayer networks (2014, methodology): “These methods have been used to extract communities [104], to rank nodes [109,110] and to perform data mining [103,111,112].”
Learning optimal ranking with tensor factorization for tag recommendation (2009, methodology): “For this minimization task, one can use standard HOSVD or square-loss implementations like [6, 8], because the data Y is assumed to be dense.”
A classification for community discovery methods in complex networks (2011, methodology): “To the best of our knowledge, only recently have some memory and time efficient techniques been developed, such as [114].”
MACH: Fast Randomized Tensor Decompositions (2009, background): “, few non zero elements as observed in [23], there exist real world problems where the tensor is dense.”
PEGASUS: A Peta-Scale Graph Mining System Implementation and Observations (2009, background): “We run PEGASUS in M45 HADOOP cluster by Yahoo! and our own cluster composed of 9 machines.”
Anomaly detection in dynamic networks: a survey (2015, methodology): “To address the intermediate blowup problem—when the input tensor and output tensors exceed memory capacity during the computation,—Memory-Efficient Tucker (MET) decomposition was proposed [46], based on the original Tucker decomposition [83].”
Tensors for Data Mining and Data Fusion (2016, methodology): “Kolda and Sun [2008] model the DBLP collaboration network as a tensor of (author, keyword, conference) and use the Tucker decomposition to identify groups of authors who publish on similar topics and on similar conferences.”
Smart Infrastructure Maintenance Using Incremental Tensor Analysis: Extended Abstract (2017, background): “The results by CP are unique provided that we permute the rank-one components [16].”
Exact Tensor Completion Using t-SVD (2015, background): “Recovery of tensors under limited number of measurements is an important problem, which arises in a variety of applications, such as recommendation systems [4], dimensionality reduction [5], multi-class learning [6],data mining [7], [8] and computer vision [9], [2].”
GigaTensor: scaling tensor analysis up by 100 times - algorithms and discoveries (2012, methodology): “In [4, 22], efficient ways of computing tensor decompositions, when the tensor is very sparse, are introduced and are implemented in the Tensor Toolbox.”
Tensor-Based Formulation and Nuclear Norm Regularization for Multienergy Computed Tomography (2013, methodology): “To date, tensor decomposition tools such CANDECOMP/PARAFAC (CP) and [24], Tucker [25] decompositions have found application in data mining and analysis for chemistry, neuroscience, computer vision and communications [26]– [28].”
SUSTain: Scalable Unsupervised Scoring for Tensors and its Application to Phenotyping (2018, background): “They have been applied successfully in diverse applications, including social network analysis [31], text mining [14], image processing [35], recommendation systems [26], brain data analysis [2] and healthcare analytics [22], to name a few.”