Abstract

Tensor decomposition operation is the basis for many data analysis tasks from clustering, trend detection, anomaly detection, to correlation analysis. One key problem with tensor decomposition, however, is its computational complexity - especially for dense data sets, the decomposition process takes exponential time in the number of tensor modes; the process is relatively faster for sparse tensors, but decomposition is still a major bottleneck in many applications. While it is possible to reduce the decomposition time by trading performance with decomposition accuracy, a drop in accuracy may not always be acceptable. In this paper, we first recognize that in many applications, the user may have a focus of interest - i.e., part of the data for which the user needs high accuracy- and beyond this area focus, accuracy may not be as critical. Relying on this observation, we propose a novel Personalized Tensor Decomposition (PTD) mechanism for accounting for the user's focus: PTD takes as input one or more areas of focus and performs the decomposition in such a way that, when reconstructed, the accuracy of the tensor is boosted for these of focus. We discuss alternative ways PTD can be implemented. Experiments show that PTD helps boost accuracy at the foci of interest, while reducing the overall tensor decomposition time.

Original languageEnglish (US)
Title of host publicationCIKM 2014 - Proceedings of the 2014 ACM International Conference on Information and Knowledge Management
PublisherAssociation for Computing Machinery, Inc
Pages689-698
Number of pages10
ISBN (Print)9781450325981
DOIs
StatePublished - Nov 3 2014
Event23rd ACM International Conference on Information and Knowledge Management, CIKM 2014 - Shanghai, China
Duration: Nov 3 2014Nov 7 2014

Other

Other23rd ACM International Conference on Information and Knowledge Management, CIKM 2014
CountryChina
CityShanghai
Period11/3/1411/7/14

Fingerprint

Tensors
Decomposition
Computational complexity

ASJC Scopus subject areas

  • Information Systems and Management
  • Computer Science Applications
  • Information Systems

Cite this

Li, X., Huang, S., Candan, K., & Sapino, M. L. (2014). Focusing decomposition accuracy by personalizing tensor decomposition (PTD). In CIKM 2014 - Proceedings of the 2014 ACM International Conference on Information and Knowledge Management (pp. 689-698). Association for Computing Machinery, Inc. https://doi.org/10.1145/2661829.2662051

Focusing decomposition accuracy by personalizing tensor decomposition (PTD). / Li, Xinsheng; Huang, Shengyu; Candan, Kasim; Sapino, Maria Luisa.

CIKM 2014 - Proceedings of the 2014 ACM International Conference on Information and Knowledge Management. Association for Computing Machinery, Inc, 2014. p. 689-698.

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

Li, X, Huang, S, Candan, K & Sapino, ML 2014, Focusing decomposition accuracy by personalizing tensor decomposition (PTD). in CIKM 2014 - Proceedings of the 2014 ACM International Conference on Information and Knowledge Management. Association for Computing Machinery, Inc, pp. 689-698, 23rd ACM International Conference on Information and Knowledge Management, CIKM 2014, Shanghai, China, 11/3/14. https://doi.org/10.1145/2661829.2662051
Li X, Huang S, Candan K, Sapino ML. Focusing decomposition accuracy by personalizing tensor decomposition (PTD). In CIKM 2014 - Proceedings of the 2014 ACM International Conference on Information and Knowledge Management. Association for Computing Machinery, Inc. 2014. p. 689-698 https://doi.org/10.1145/2661829.2662051
Li, Xinsheng ; Huang, Shengyu ; Candan, Kasim ; Sapino, Maria Luisa. / Focusing decomposition accuracy by personalizing tensor decomposition (PTD). CIKM 2014 - Proceedings of the 2014 ACM International Conference on Information and Knowledge Management. Association for Computing Machinery, Inc, 2014. pp. 689-698
@inproceedings{9c44d784f3f44bdcb656348e526eed10,
title = "Focusing decomposition accuracy by personalizing tensor decomposition (PTD)",
abstract = "Tensor decomposition operation is the basis for many data analysis tasks from clustering, trend detection, anomaly detection, to correlation analysis. One key problem with tensor decomposition, however, is its computational complexity - especially for dense data sets, the decomposition process takes exponential time in the number of tensor modes; the process is relatively faster for sparse tensors, but decomposition is still a major bottleneck in many applications. While it is possible to reduce the decomposition time by trading performance with decomposition accuracy, a drop in accuracy may not always be acceptable. In this paper, we first recognize that in many applications, the user may have a focus of interest - i.e., part of the data for which the user needs high accuracy- and beyond this area focus, accuracy may not be as critical. Relying on this observation, we propose a novel Personalized Tensor Decomposition (PTD) mechanism for accounting for the user's focus: PTD takes as input one or more areas of focus and performs the decomposition in such a way that, when reconstructed, the accuracy of the tensor is boosted for these of focus. We discuss alternative ways PTD can be implemented. Experiments show that PTD helps boost accuracy at the foci of interest, while reducing the overall tensor decomposition time.",
author = "Xinsheng Li and Shengyu Huang and Kasim Candan and Sapino, {Maria Luisa}",
year = "2014",
month = "11",
day = "3",
doi = "10.1145/2661829.2662051",
language = "English (US)",
isbn = "9781450325981",
pages = "689--698",
booktitle = "CIKM 2014 - Proceedings of the 2014 ACM International Conference on Information and Knowledge Management",
publisher = "Association for Computing Machinery, Inc",

}

TY - GEN

T1 - Focusing decomposition accuracy by personalizing tensor decomposition (PTD)

AU - Li, Xinsheng

AU - Huang, Shengyu

AU - Candan, Kasim

AU - Sapino, Maria Luisa

PY - 2014/11/3

Y1 - 2014/11/3

N2 - Tensor decomposition operation is the basis for many data analysis tasks from clustering, trend detection, anomaly detection, to correlation analysis. One key problem with tensor decomposition, however, is its computational complexity - especially for dense data sets, the decomposition process takes exponential time in the number of tensor modes; the process is relatively faster for sparse tensors, but decomposition is still a major bottleneck in many applications. While it is possible to reduce the decomposition time by trading performance with decomposition accuracy, a drop in accuracy may not always be acceptable. In this paper, we first recognize that in many applications, the user may have a focus of interest - i.e., part of the data for which the user needs high accuracy- and beyond this area focus, accuracy may not be as critical. Relying on this observation, we propose a novel Personalized Tensor Decomposition (PTD) mechanism for accounting for the user's focus: PTD takes as input one or more areas of focus and performs the decomposition in such a way that, when reconstructed, the accuracy of the tensor is boosted for these of focus. We discuss alternative ways PTD can be implemented. Experiments show that PTD helps boost accuracy at the foci of interest, while reducing the overall tensor decomposition time.

AB - Tensor decomposition operation is the basis for many data analysis tasks from clustering, trend detection, anomaly detection, to correlation analysis. One key problem with tensor decomposition, however, is its computational complexity - especially for dense data sets, the decomposition process takes exponential time in the number of tensor modes; the process is relatively faster for sparse tensors, but decomposition is still a major bottleneck in many applications. While it is possible to reduce the decomposition time by trading performance with decomposition accuracy, a drop in accuracy may not always be acceptable. In this paper, we first recognize that in many applications, the user may have a focus of interest - i.e., part of the data for which the user needs high accuracy- and beyond this area focus, accuracy may not be as critical. Relying on this observation, we propose a novel Personalized Tensor Decomposition (PTD) mechanism for accounting for the user's focus: PTD takes as input one or more areas of focus and performs the decomposition in such a way that, when reconstructed, the accuracy of the tensor is boosted for these of focus. We discuss alternative ways PTD can be implemented. Experiments show that PTD helps boost accuracy at the foci of interest, while reducing the overall tensor decomposition time.

UR - http://www.scopus.com/inward/record.url?scp=84937546380&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84937546380&partnerID=8YFLogxK

U2 - 10.1145/2661829.2662051

DO - 10.1145/2661829.2662051

M3 - Conference contribution

AN - SCOPUS:84937546380

SN - 9781450325981

SP - 689

EP - 698

BT - CIKM 2014 - Proceedings of the 2014 ACM International Conference on Information and Knowledge Management

PB - Association for Computing Machinery, Inc

ER -