Approximate tensor decomposition within a tensor-relational algebraic framework

Mijung Kim, Kasim Candan

Research output: Chapter in Book/Report/Conference proceedingConference contribution

12 Citations (Scopus)

Abstract

In this paper, we first introduce a tensor-based relational data model and define algebraic operations on this model. We note that, while in traditional relational algebraic systems the join operation tends to be the costliest operation of all, in the tensor-relational framework presented here, tensor decomposition becomes the computationally costliest operation. Therefore, we consider optimization of tensor decomposition operations within a relational algebraic framework. This leads to a highly efficient, effective, and easy-to-parallelize join-by-decomposition approach and a corresponding KL-divergence based optimization strategy. Experimental results provide evidence that minimizing KL-divergence within the proposed join-by-decomposition helps approximate the conventional join-then-decompose scheme well, without the associated time and space costs.

Original languageEnglish (US)
Title of host publicationInternational Conference on Information and Knowledge Management, Proceedings
Pages1737-1742
Number of pages6
DOIs
StatePublished - 2011
Event20th ACM Conference on Information and Knowledge Management, CIKM'11 - Glasgow, United Kingdom
Duration: Oct 24 2011Oct 28 2011

Other

Other20th ACM Conference on Information and Knowledge Management, CIKM'11
CountryUnited Kingdom
CityGlasgow
Period10/24/1110/28/11

Fingerprint

Decomposition
Join
Divergence
Costs

Keywords

  • approximate tensor decomposition
  • tensor relational algebra
  • tensor-based relational model

ASJC Scopus subject areas

  • Business, Management and Accounting(all)
  • Decision Sciences(all)

Cite this

Kim, M., & Candan, K. (2011). Approximate tensor decomposition within a tensor-relational algebraic framework. In International Conference on Information and Knowledge Management, Proceedings (pp. 1737-1742) https://doi.org/10.1145/2063576.2063827

Approximate tensor decomposition within a tensor-relational algebraic framework. / Kim, Mijung; Candan, Kasim.

International Conference on Information and Knowledge Management, Proceedings. 2011. p. 1737-1742.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kim, M & Candan, K 2011, Approximate tensor decomposition within a tensor-relational algebraic framework. in International Conference on Information and Knowledge Management, Proceedings. pp. 1737-1742, 20th ACM Conference on Information and Knowledge Management, CIKM'11, Glasgow, United Kingdom, 10/24/11. https://doi.org/10.1145/2063576.2063827
Kim M, Candan K. Approximate tensor decomposition within a tensor-relational algebraic framework. In International Conference on Information and Knowledge Management, Proceedings. 2011. p. 1737-1742 https://doi.org/10.1145/2063576.2063827
Kim, Mijung ; Candan, Kasim. / Approximate tensor decomposition within a tensor-relational algebraic framework. International Conference on Information and Knowledge Management, Proceedings. 2011. pp. 1737-1742
@inproceedings{ad33b5bb6d82435cbf81dec00885a943,
title = "Approximate tensor decomposition within a tensor-relational algebraic framework",
abstract = "In this paper, we first introduce a tensor-based relational data model and define algebraic operations on this model. We note that, while in traditional relational algebraic systems the join operation tends to be the costliest operation of all, in the tensor-relational framework presented here, tensor decomposition becomes the computationally costliest operation. Therefore, we consider optimization of tensor decomposition operations within a relational algebraic framework. This leads to a highly efficient, effective, and easy-to-parallelize join-by-decomposition approach and a corresponding KL-divergence based optimization strategy. Experimental results provide evidence that minimizing KL-divergence within the proposed join-by-decomposition helps approximate the conventional join-then-decompose scheme well, without the associated time and space costs.",
keywords = "approximate tensor decomposition, tensor relational algebra, tensor-based relational model",
author = "Mijung Kim and Kasim Candan",
year = "2011",
doi = "10.1145/2063576.2063827",
language = "English (US)",
isbn = "9781450307178",
pages = "1737--1742",
booktitle = "International Conference on Information and Knowledge Management, Proceedings",

}

TY - GEN

T1 - Approximate tensor decomposition within a tensor-relational algebraic framework

AU - Kim, Mijung

AU - Candan, Kasim

PY - 2011

Y1 - 2011

N2 - In this paper, we first introduce a tensor-based relational data model and define algebraic operations on this model. We note that, while in traditional relational algebraic systems the join operation tends to be the costliest operation of all, in the tensor-relational framework presented here, tensor decomposition becomes the computationally costliest operation. Therefore, we consider optimization of tensor decomposition operations within a relational algebraic framework. This leads to a highly efficient, effective, and easy-to-parallelize join-by-decomposition approach and a corresponding KL-divergence based optimization strategy. Experimental results provide evidence that minimizing KL-divergence within the proposed join-by-decomposition helps approximate the conventional join-then-decompose scheme well, without the associated time and space costs.

AB - In this paper, we first introduce a tensor-based relational data model and define algebraic operations on this model. We note that, while in traditional relational algebraic systems the join operation tends to be the costliest operation of all, in the tensor-relational framework presented here, tensor decomposition becomes the computationally costliest operation. Therefore, we consider optimization of tensor decomposition operations within a relational algebraic framework. This leads to a highly efficient, effective, and easy-to-parallelize join-by-decomposition approach and a corresponding KL-divergence based optimization strategy. Experimental results provide evidence that minimizing KL-divergence within the proposed join-by-decomposition helps approximate the conventional join-then-decompose scheme well, without the associated time and space costs.

KW - approximate tensor decomposition

KW - tensor relational algebra

KW - tensor-based relational model

UR - http://www.scopus.com/inward/record.url?scp=83055166026&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=83055166026&partnerID=8YFLogxK

U2 - 10.1145/2063576.2063827

DO - 10.1145/2063576.2063827

M3 - Conference contribution

AN - SCOPUS:83055166026

SN - 9781450307178

SP - 1737

EP - 1742

BT - International Conference on Information and Knowledge Management, Proceedings

ER -