home Forums # Technical Support hello Juan on fuzzylite aid for students‏

Viewing 2 posts - 1 through 2 (of 2 total)
  • Author
    Posts
  • #1480
    Unknown
    Member

    Greetings Juan Rada, I am developing an expert system for diagnosis of pests and diseases on rice. This is based on graduation thesis, I would like to know how JFuzzyLite, gives a result for example the size of a male insect is 2.50 to 3.10mm and a female of 2.90 to 3.80 but what distinguishes the two is the color, such rules would

    IF INSECT IS BROWN AND SIZE> = 2.50 AND SIZE <= 3.10 IS THEN INSECT CIGARRITA_MACHO
    ELSE IF COLOR IS BLACK AND SIZE>=2.90 AND SIZE<=3.80 THEN INSECT CIGARRITA_HEMBRA

    The question is whether JFUZZYLITE GIVE ME COULD RESULT “CIGARRITA_MACHO” or the result given in numbers, could explain or would have an example. Thank you very much in advance.

    #1483

    Hi Roberth,

    fuzzylite deals with uncertainty. The type of rules would be more like:

    if colour is light and size is small then insect is cigarrita_macho
    if colour is dark and size is big then insect is cigarrita_hembra

    The output will be a degree of “confidence” with which the insect is cigarrita_hembra and _macho, e.g., 0.8/hembra +0.2/macho. You could also defuzzify this output and obtain a decimal value in the range of defition of insect. For example, if insect is defined in [0.0,1.0], the defuzzification could be 0.8333.

    Please, revise the examples in QtFuzzyLite.

    I hope this helps.

    Cheers,

Viewing 2 posts - 1 through 2 (of 2 total)
  • You must be logged in to reply to this topic.