An adaptive neuro-fuzzy inference system (ANFIS) is a fuzzy system whose membership function parameters have been tuned using neuro-adaptive learning methods similar to those used in training neural networks. Fuzzy Logic Toolbox™ software provides command-line functions and an app for training Sugeno-type fuzzy inference systems using given input/output training data. For more information see Neuro-Adaptive Learning and ANFIS.
|Neuro-Fuzzy Designer||Design, train, and test Sugeno-type fuzzy inference systems|
||Create new Fuzzy Inference System|
||Generate Fuzzy Inference System structure from data using grid partition|
||Generate Fuzzy Inference System structure from data using subtractive clustering|
||Generate Fuzzy Inference System structure from data using FCM clustering|
||Transform Mamdani Fuzzy Inference System into Sugeno Fuzzy Inference System|
Fuzzy inference maps an input space to an output space using a series of fuzzy if-then rules.
In Sugeno systems, the output of each if-then rule is either constant or a linear function of the input variables. The final output value is the weighted average of all rule outputs.
You can tune Sugeno fuzzy inference systems using neuro-adaptive learning techniques similar to those used for training neural networks.
You can design neuro-fuzzy systems either at the command-line or using the Neuro-Fuzzy Designer app.
Interactively create, train, and test neuro-fuzzy systems using the Neuro-Fuzzy Designer app.
Validate trained neuro-fuzzy systems using checking data that is different from training data.
When using Neuro-Fuzzy Designer, you can export your trained neuro-fuzzy model and training error data to the MATLAB® workspace for further analysis.