• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to previous page

Embedding time series data in Euclidean space from DTW distances

Research Project

Project/Area Number 18500116
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeSingle-year Grants
Section一般
Research Field Intelligent informatics
Research InstitutionHiroshima City University

Principal Investigator

HAYASHI Akira  Hiroshima City University, Faculty of Information Sciences, Professor (60240909)

Co-Investigator(Kenkyū-buntansha) SUEMATSU Nobuo  Hiroshima City University, Faculty of Information Sciences, Associate Professor (70264942)
Project Period (FY) 2006 – 2007
Project Status Completed (Fiscal Year 2007)
Budget Amount *help
¥3,380,000 (Direct Cost: ¥2,900,000、Indirect Cost: ¥480,000)
Fiscal Year 2007: ¥2,080,000 (Direct Cost: ¥1,600,000、Indirect Cost: ¥480,000)
Fiscal Year 2006: ¥1,300,000 (Direct Cost: ¥1,300,000)
Keywordstime series data / embedding / multi-dimensional scaling / dynamic time warping / kernel method / semidefinite proaramming / time series data / dynamic time warping / embedding / multidimensional scaling / semidefinite programming
Research Abstract

One of the advantages of the kernel methods is that they can deal with various kinds of objects, not necessarily vectorial data with a fixed number of attributes.
In this paper, we develop kernels for time series data using dynamic time warping (DTW) distances. Since DTW distances are pseudo distances that do not satisfy the triangle inequality, a kernel matrix based on them is not positive semidefinite, in general. We use semidefinite programming (SDP) to guarantee the positive definiteness of a kernel matrix. We present neighborhood preserving embedding (NPE), an SDP formulation to obtain a kernel matrix that best preserves the local geometry of time series data. We also present an out-of-sample extension (OSE) for NPE.
We use two applications, time series classification and time series embedding for similarity search to validate our approach.

Report

(3 results)
  • 2007 Annual Research Report   Final Research Report Summary
  • 2006 Annual Research Report
  • Research Products

    (13 results)

All 2008 2007 2006 Other

All Journal Article (10 results) (of which Peer Reviewed: 4 results) Presentation (3 results)

  • [Journal Article] 層的混合モデル学習のためのコンポーネント削減法2008

    • Author(s)
      前橋 久美子
    • Journal Title

      電子情報通信学会論文誌 J-91D(4)

      Pages: 1058-1068

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2007 Annual Research Report 2007 Final Research Report Summary
    • Peer Reviewed
  • [Journal Article] A Redundancy-Based Measure of Dissimilarity among Probability Distributions for Hierarchical Clustering Criteria2008

    • Author(s)
      Kazunori Iwata
    • Journal Title

      IEEE Transactions on Pattern Analysis and Machine Intelligence 30(1)

      Pages: 76-88

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2007 Annual Research Report 2007 Final Research Report Summary
    • Peer Reviewed
  • [Journal Article] Reduction in the number of components in mixture models2008

    • Author(s)
      Kumiko, Maebashi, Nobuo, Suematsu, Akira, Hayashi
    • Journal Title

      The IEICE Transactions on Information and Systems(Japanese Edition) J-91D(4)

      Pages: 1058-1068

    • NAID

      110007381051

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2007 Final Research Report Summary
  • [Journal Article] A Redundancy-Based Measure of Dissimilarity among Probability Distributions for Hierarchical Clustering Criteria2008

    • Author(s)
      Kazunori, Iwata, Akira, Hayashi
    • Journal Title

      IEEE Transactions on Pattern Analysis and Machine Intelligence 30(1)

      Pages: 76-88

    • NAID

      120005402653

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2007 Final Research Report Summary
  • [Journal Article] 確率密度推定に基づくRDSP法を用いた音素データの階層クラスタ分析2008

    • Author(s)
      斧城 悠大
    • Journal Title

      電子情報通信学会論文誌D J-91D(8)(掲載確定)

    • Related Report
      2007 Annual Research Report
    • Peer Reviewed
  • [Journal Article] A Discriminative Model Corresponding to Hierarchical HMMs2007

    • Author(s)
      T.Sugiura
    • Journal Title

      Lecture Note in Computer Science 4881

      Pages: 374-385

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2007 Final Research Report Summary
    • Peer Reviewed
  • [Journal Article] A Discriminative Model Corresponding to Hierarchical HMMs2007

    • Author(s)
      Takaaki, Sugiura, Naoto, Gotou, Akira, Hayashi
    • Journal Title

      Lecture Note in Computer Science 4881

      Pages: 375-384

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2007 Final Research Report Summary
  • [Journal Article] Theory of a Probabilistic-Dependence Measure of dissimilarity among Multiple Clusters2006

    • Author(s)
      Kazunori Iwata, Akira Hayashi
    • Journal Title

      Lecture Notes in Computer Science 4132

      Pages: 311-320

    • Related Report
      2006 Annual Research Report
  • [Journal Article] パーティクルフィルタを用いた歩き方による個人識別2006

    • Author(s)
      江本光晴, 林朗, 末松伸朗, 岩田一貴
    • Journal Title

      信学技報 106(230)

      Pages: 41-48

    • NAID

      110004820688

    • Related Report
      2006 Annual Research Report
  • [Journal Article] A Redundancy-Based Measure of Dissimilarity among Probability Distributions for Hierarchical Clustering Criteria

    • Author(s)
      Kazunori Iwata, Akira Hayashi
    • Journal Title

      IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI) Accepted

    • NAID

      120005402653

    • Related Report
      2006 Annual Research Report
  • [Presentation] Learning a kernel matrix for time series data2007

    • Author(s)
      Hiroyuki Narita
    • Organizer
      Proc.ICONIP 2007 14th International Conference Neural Information Processing
    • Place of Presentation
      北九州市
    • Year and Date
      2007-11-15
    • Related Report
      2007 Annual Research Report
  • [Presentation] Learning a kernel matrix for time series data2007

    • Author(s)
      Hiroyuki Narita
    • Organizer
      Proc.ICONIP 2007 14th International Conference on Neural Information Processing
    • Place of Presentation
      北九州市
    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2007 Final Research Report Summary
  • [Presentation] Learning a kernel matrix for time series data2007

    • Author(s)
      Hiroyuki, Narita, Yasumasa, Sawamura, Akira, Hayashi
    • Organizer
      Proc. ICONIP 2007 14th International Conference on Neural Information Processing
    • Place of Presentation
      Kitakyusyu, Japan
    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2007 Final Research Report Summary

URL: 

Published: 2006-04-01   Modified: 2016-04-21  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi