Fuzzy Relations

[Unknown]

Hi Juan,
Is it possible to use fuzzy relations with the library ?
Example: Velocity = Diameter ^ 0.15 (Velocity and Diameter are variables, membership not a crisp relation)

If possible, can you please give a simple example for qtfuzzylite (for beginners in fuzzy) ?

Thank you in advance.

Regards

Thomas

[Juan Rada-Vilela (admin)]

Hi Thomas.

Thank you for your post.

In fuzzylite, you have a Function membership function, which takes any string and converts it into a binary expression tree and evaluates the expression at runtime. You can have any relationship between variables you want. You can check it out in QtFuzzyLite 5.

Cheers.

[Unknown]

Hi Juan,
Thank you very much for your support.
I found the term “function” in QtFuzzyLite 4, but all attempts brought no satisfactory results.

Regards

[Juan Rada-Vilela (admin)]

Hi Thomas,

I think fuzzylite 4 had a funny parser of expressions for Function. What do you mean exactly by not satisfactory? What is wrong with Function in QtFL4?

[Unknown]

Hi Juan,

yes the parser can have many expressions. I do not think that there is something wrong with the parser.
My Problem is that I want to have a fuzzy relation of 2 parameters (here: diameter + velocity) to get the mebership of a precipitation type (e.g. rain, should look like https://www.researchgate.net/figure/50292056_fig7_Fig-10-Constructed-fuzzy-relations-on-the-crossbar-of-the-circuit-of-Fig-6-based-on).

Cheers,

[Juan Rada-Vilela (admin)]

Hi Thomas,

you are going to need to provide more information about your requirements. Could you please provide a fuzzylite engine and present what you wish to achieve?

You can create any number of linguistic variables you want, and refer to their values from any other variable in a Function membership function using any number of mathematical operations, but I am not sure if this is what you are aiming for. I would need more information to help you.

Cheers.

[Unknown]

Hi Juan,

now we have decided not to use fuzzylite since we have apparently not sufficient knowhow as fuzzy beginners. We will perform our calculations as in the past with standard logic.

Thank you for your efforts.

Cheers.

[Author]

Posts