'not' and 'any' about help HOT 2 CLOSED

Hi,

Thanks for your question.

any produces a membership function value of 1 always. I think any is better suited for output variables.

However, I understand what you mean and expected.

Also, if the value of ELE does not matter, maybe the rule is better and simpler by omitting that term?

Does this answer your question?

from help.

Hi Mr Vileda

Thank you for your software and for the response.

As you know (see my E-mail), I try to get a value (output) of aerial objects distance (DIS) with testimonial values, and words of appreciations of angular size (TAG), elevation (ELE), light intensities (LUM), speed (VIT) and sound (BRT). This is for UFO study. In many cases the data are not complete, or not defined (no elevation, no speed appreciation...)

I' think I've solved my problem for 'not' . I keep only "TAG 'Petite' and ELE... then ...". But I have added two more rules to get the expected kind of output. My personal "logic" says that 'not' is not these specific function, then all the others and not 'nan' if the input variable is not defined; and with 'any' I understand all the defined values and then all the possible functions without 'nan'.

ELE must be present in any case with TAG following the study of literature about human perception, with one exception, if sound is audible. I think that I have now a "stable" model if I give the value 0 in case of ELE and all the others variables (except TAG) are 'nan'. I have started TAG and ELE at 1. Then if the value is 0 there is no fuzzy function associated with.

Bye.

from help.