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

非負値テンソル因子化法を用いた多次元・多関係データ解析手法の研究

Research Project

Project/Area Number 14J01495
Research Category

Grant-in-Aid for JSPS Fellows

Allocation TypeSingle-year Grants
Section国内
Research Field Mathematical informatics
Research InstitutionHokkaido University

Principal Investigator

木村 圭吾  北海道大学, 情報科学研究科, 特別研究員(DC1)

Project Period (FY) 2014-04-25 – 2017-03-31
Project Status Completed (Fiscal Year 2016)
Budget Amount *help
¥1,200,000 (Direct Cost: ¥1,200,000)
Fiscal Year 2016: ¥400,000 (Direct Cost: ¥400,000)
Fiscal Year 2015: ¥400,000 (Direct Cost: ¥400,000)
Fiscal Year 2014: ¥400,000 (Direct Cost: ¥400,000)
Keywords非負値テンソル分解 / 非負値行列分解 / マルチラベル分類 / Online NTF
Outline of Annual Research Achievements

Nonnegative Tensor Factorization, Nonnegative Matrix Factorizationはテンソル,行列津で表現されたデータを人間に解釈が容易である非負で疎な成分に分解する手法であり,これらによってデータの潜在的なパターンを抽出することに広く用いられている手法である.本研究の目的は,NTF/NMFの高速アルゴリズム及び,オンラインアルゴリズムの提案,そしてそれらの応用である.
今年度は主に昨年度までに提案した手法の関連手法や関連分野での位置づけを行った.特に今までにNonnegative Matrix Factorization及びNonnegative Tensor Factorizationがどのような分野でどのように応用として扱われたかをまとめ,提案法が有効な分野とその応用先についての模索を行った.これらは博士論文という形で執筆を行った.
前年度から続けている応用先のマルチラベル分類手法はInternational Conference on Pattern Recognition (ICPR)およびStructural and Syntactic Pattern Recognition and Statistical Techniques in Pattern Recognition (S+SSPR)に採録された.
これらの手法は現在共著の学生と共にMATLAB上で動くライブラリ開発を行っている状態で,論文の投稿を含めて今後も活動を行っていく予定である.

Research Progress Status

28年度が最終年度であるため、記入しない。

Strategy for Future Research Activity

28年度が最終年度であるため、記入しない。

Report

(3 results)
  • 2016 Annual Research Report
  • 2015 Annual Research Report
  • 2014 Annual Research Report
  • Research Products

    (8 results)

All 2017 2016 2015 2014 Other

All Journal Article (1 results) (of which Peer Reviewed: 1 results) Presentation (5 results) (of which Int'l Joint Research: 4 results) Remarks (2 results)

  • [Journal Article] A Column-wise Update Algorithm for Nonnegative Matrix Factorization in Bregman Divergence with Orthogonal Constraint2016

    • Author(s)
      Keigo Kimura, Mineichi Kudo and Yuzuru Tanaka
    • Journal Title

      Machine Learning

      Volume: - Issue: 2 Pages: 1-22

    • DOI

      10.1007/s10994-016-5553-0

    • NAID

      120006220037

    • Related Report
      2015 Annual Research Report
    • Peer Reviewed
  • [Presentation] Fast Random k-labelsets for Large-Scale Multi-Label Classification2017

    • Author(s)
      Keigo Kimura
    • Organizer
      International Conference on Pattern Recognition (ICPR 2016)
    • Place of Presentation
      Cancun, Mexico
    • Year and Date
      2017-12-04
    • Related Report
      2016 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Simultaneous Nonliner Label-Instance Embedding for Multi-Label Classification2017

    • Author(s)
      Keigo Kimura
    • Organizer
      Joint IAPR International Workshops on Statistical Techniques in Pattern Recognition (SPR) and Structural and Syntactic Pattern Recognition (SSPR)
    • Place of Presentation
      Merida, Mexico
    • Year and Date
      2017-11-30
    • Related Report
      2016 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Variable Selection for Efficient Nonnegative Tensor Factorization.2015

    • Author(s)
      Keigo Kimura
    • Organizer
      the 2015 IEEE International Conference on Data Mining (ICDM)
    • Place of Presentation
      Atlantic City, USA
    • Year and Date
      2015-11-14
    • Related Report
      2015 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Dimension Reduction Using Nonnegative Matrix Tri-Factorization in Multi-label Classification2015

    • Author(s)
      Keigo Kimura
    • Organizer
      the 2015 International Conference on Parallel & Distributed Processing Techniques & Applications: Workshop on Mathematical Modeling and Problem Solving
    • Place of Presentation
      Las Vegas, USA
    • Year and Date
      2015-07-27
    • Related Report
      2015 Annual Research Report
    • Int'l Joint Research
  • [Presentation] A Fas Hierarchical Alternating Least Squares Algorithm for Orthogonal Nonnegative Matrix Factorization2014

    • Author(s)
      Keigo Kimura
    • Organizer
      The 6th Asian Conference on Machine Learning 2014
    • Place of Presentation
      Nha Trang, Vietnam,
    • Year and Date
      2014-11-26 – 2014-11-28
    • Related Report
      2014 Annual Research Report
  • [Remarks] 個人のホームページ

    • URL

      https://sites.google.com/site/kkimura360/home

    • Related Report
      2015 Annual Research Report
  • [Remarks] 著者について

    • URL

      http://prml.main.ist.hokudai.ac.jp/~kkimura/

    • Related Report
      2014 Annual Research Report

URL: 

Published: 2015-01-22   Modified: 2024-03-26  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi