Akademska digitalna zbirka SLovenije - logo
ALL libraries (COBIB.SI union bibliographic/catalogue database)
  • Decision trees with use fuzzy heuristics function
    Hleb Babič, Špela ; Kokol, Peter, 1957-
    Decision trees are relative known and often used intelligent tool for decision support. They are convenient for capturing the knowledge from vast amount of data. The result or decision model is ... represented in hierarchical manner, where the significance or contribution of single attribute to final decision is shown very clearlt. When the decision tree is built upon real-world data this data is more often numeric than descrete. While it is in human nature to better use the words than numbers in order to describe somenthing the fuzzy logic theory was introduced to fill this gap. Description of the attribute properties using fuzzy logic approach is represented by vector of plausibility where the coordinates between 0 and 1 show the plausibility of attribute to its subsets of possible attribute values. This is the way to successfully overcome the problem of boundary values between attribute subsets where the sharp determined boundaries can incorrectly affect the final result a lot. In Laboratory for system design we have developed the software tool for building the decision trees with fuzzy heurisctics function and using them for decision support in everyday practice in medical environments.
    Type of material - conference contribution
    Publish date - 2000
    Language - english
    COBISS.SI-ID - 5621526