逐次的なユーザ・アイテムクラスタ抽出に基づく協調フィルタリング  [in Japanese] Collaborative Filtering Based on Sequential Extraction of User-Item Clusters  [in Japanese]

Access this Article

Search this Article

Author(s)

Abstract

Collaborative filtering is a computational realization of “word-of-mouth” in network community, in which the items prefered by “neighbors” are recommended. This paper proposes a new item-selection model for extracting user-item clusters from rectangular relation matrices, in which mutual relations between users and items are denoted in an alternative process of “liking or not”. A technique for sequential co-cluster extraction from rectangular relational data is given by combining the structural balancing-based user-item clustering method with sequential fuzzy cluster extraction appraoch. Then, the tecunique is applied to the collaborative filtering problem, in which some items may be shared by several user clusters. <br>

Journal

  • Transactions of the Institute of Systems, Control and Information Engineers

    Transactions of the Institute of Systems, Control and Information Engineers 22(10), 364-370, 2009-10-15

    THE INSTITUTE OF SYSTEMS, CONTROL AND INFORMATION ENGINEERS (ISCIE)

References:  21

Cited by:  1

Codes

  • NII Article ID (NAID)
    10026193995
  • NII NACSIS-CAT ID (NCID)
    AN1013280X
  • Text Lang
    JPN
  • Article Type
    Journal Article
  • ISSN
    13425668
  • NDL Article ID
    10383958
  • NDL Source Classification
    ZM11(科学技術--科学技術一般--制御工学)
  • NDL Call No.
    Z14-195
  • Data Source
    CJP  CJPref  NDL  J-STAGE 
Page Top