• DocumentCode
    2115647
  • Title

    Intelligent Classification Based on Tolerance Relations

  • Author

    Zhou, Pei Li ; Cai, Wanjing

  • Author_Institution
    Fac. of Inf. Technol., Monash Univ., Clayton, VIC
  • fYear
    2008
  • fDate
    18-18 Dec. 2008
  • Firstpage
    445
  • Lastpage
    448
  • Abstract
    Binary relations and its operations are important in set theory. But traditional research mainly focuses on pure mathematics operations on set theory. Pansystems strengthen the research work in set theory by combining mathematics with its philosophical background. In this paper, using pansystems PMT-combination method, we re-inspect the philosophical background of binary relations and give a new interpretation of equivalence relations and tolerance relations. Tolerance relations are more generic than equivalence relations. Equivalence relations can generate a division on a given set. Tolerance relations, however, cannot to be used to divide a set directly. Based on our research on tolerance relations and equivalence relations, we propose a new method of using tolerance relations to divide a given set. Furthermore we go back to the real world and give the philosophical explanation of this new kind of division.
  • Keywords
    equivalence classes; learning (artificial intelligence); pattern classification; set theory; PMT-combination method; binary relation; equivalence relation; intelligent classification; machine learning; mathematics operation; pansystem; set theory; tolerance relation; Abstracts; Biomedical engineering; Books; Computer science; Cybernetics; Information technology; Mathematics; Pediatrics; Seminars; Set theory; Binary Relations; Equivalence Relations; Intelligence Classfication; Pansystems; Tolerance Relations;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Future BioMedical Information Engineering, 2008. FBIE '08. International Seminar on
  • Conference_Location
    Wuhan, Hubei
  • Print_ISBN
    978-0-7695-3561-6
  • Type

    conf

  • DOI
    10.1109/FBIE.2008.120
  • Filename
    5076778