Hey Brian, it is so wonderful to see that how you make literally anything look way simpler than they actually are! Looking forward to see the following sections!
This is great and all, but I for one already interacted with making an FIS by using it with the tip example. Would like to see real applications of FIS with AI somewhere, lol. Though I will say that so far this has been the greatest walkthrough for demonstrating the fuzzy system.
Hi Brian, thank you for your informative video! I have a question related to the graph shown in 15:32. Accross the whole service range, we can see a small dip in tip amount around food quality '8'. Is it not weird that the tip amount is slightly higher for food quality 6 than for food quality 8? I tried myself making such FIS and I wonder if there's a way of avoiding this, or should I see this as a small downside of the FIS approach?
This is kind of like a multivalued likelihood. For example, each fuzzy membership function is a possible distribution of how the values are distributed and the crisp evaluations are those likelihoods. So should we be able to do maximum likelihood estimation for our fuzzy membership functions?
Fuzzyfierr : V1 > degree of membership 1 rule based inference : Degree of membership v 1 > Degree of membership v2 Defuzzifier :Degree of membership 2 > v 2
wow, such a great explaining, thank you, one question please, I just couldn't understand why you chose a specific shape of membership for each one of the inputs? why couldn't all be gaussian for example? thank you again
Hello, thank you, very clear explanation. I just have a question: The centroid we get at defuzzification step, is that in fuzzified values for the output variable (0-1), or crisps value? Because when you plot 3d surface, all seem to be in crisp value? Thanks.
Hey Brian, it is so wonderful to see that how you make literally anything look way simpler than they actually are! Looking forward to see the following sections!
This type of work is the one that deserves support. Just to show some appreciation.
If Brian uploads video then I am happy.. ok thats hard logic
ur vd help a lot, now i no need watch lectures from uni. it covers all the concept. Thanks a lot!
Really looking forwards to the rest of this series. Super useful
Awsome! Such an amazing and intuitive video about this topic.
I love fuzzy logic, we use it for diagnoses purposes
Thank you for watching. Love to learn about your use cases.
Great Video!! I don't known the fuzzy analysis and the video it was very clear. I will try this kind of solution in some real life situation.
Very good illustration level. My special thanks ❤
crazy! tks a lot Brian, u made it simple and understandable! hat off!
Glad it helped!
This is great and all, but I for one already interacted with making an FIS by using it with the tip example. Would like to see real applications of FIS with AI somewhere, lol. Though I will say that so far this has been the greatest walkthrough for demonstrating the fuzzy system.
Exactly. I'm supposed to make multiple inputs and multiple outputs fuzzy model and TH-cam isn't being helpful :)
What a great explanation, once again!
Thank you a lot! Your videos are amazing!
Very neat explanatory, thank you
Excellent illustrations !
Best explanation
Congrats
very very cool. You are an amazing teacher brian, thank you.
Hi Brian, thank you for your informative video! I have a question related to the graph shown in 15:32. Accross the whole service range, we can see a small dip in tip amount around food quality '8'. Is it not weird that the tip amount is slightly higher for food quality 6 than for food quality 8? I tried myself making such FIS and I wonder if there's a way of avoiding this, or should I see this as a small downside of the FIS approach?
This is kind of like a multivalued likelihood. For example, each fuzzy membership function is a possible distribution of how the values are distributed and the crisp evaluations are those likelihoods.
So should we be able to do maximum likelihood estimation for our fuzzy membership functions?
so you suggest empirical verification of membership functions via polling but what about the interpretation of fuzzy operators?
Fuzzyfierr : V1 > degree of membership 1
rule based inference : Degree of membership v 1 > Degree of membership v2
Defuzzifier :Degree of membership 2 > v 2
How did the 4th rule get added to the 3rd rule to produce the new value of High? Did you connect them by OR?
Dear brian , can u provide us with the matlab codes and simulink you've shown us in the videos?
so good. loved it.
Wonderful job and just great. Thank you very much
2:25 why is is 55/45 and not 50/50? It seems like 7 is the midway point
Waiting for the next video
absoultely amazing
Fantastic content - thank's a lot!
I wish you include MATLAB example file with this excellent illustration
Nice video. Could you please provide us the fuzzy raw file you developed rather than using the toolbox?
Thank you for the amazing video, it was really helpful
can you use some fuzzy logic to teach the robovoice how to pronounce adjective & inference? such an obvious tell.
You rock man:) Keep it up.
How to find rules on the basis of clusters using fuzzy c means
Please explain
wow, such a great explaining, thank you, one question please, I just couldn't understand why you chose a specific shape of membership for each one of the inputs? why couldn't all be gaussian for example? thank you again
Can someone help me to give video link of part 1?
Thanks for the great explanation.
Is it possible to get the link for the application you used in the video?
Can this type of layout is really in matlab? Or just edited?
great explanation. thanks
hi Sir , I hope that ur fine i have question
what 's the diffirent if we use a TRAPEZOID fonction or GAUSSIENNE Fonction or TRIANGLE fonction?
Is it possible to create the model using the neuro fuzzy designer but deploy the model to html
Can you share the programs
It's a great explaination
Master piece class thanks
how to change the scale of y-axis in Fuzzy logic Designer?
Very interesting
Hello, thank you, very clear explanation. I just have a question:
The centroid we get at defuzzification step, is that in fuzzified values for the output variable (0-1), or crisps value? Because when you plot 3d surface, all seem to be in crisp value? Thanks.
I will reassure everyone that if the food is dog and the service is dog, the tip will be a nice round 0%
Bro is making tipping obligatory
It's not me! it's the inference statements combined with the membership functions 😅
great!
when is the next part expected?
How I hope if I had watched this 6 months ago ~~
This will sound odd but I promise, to somebody it makes sense: "fuzziness is next to Rodneyness"...
Translating Matlab to Spanish, please.