Previous Page

5. Adaptive Fuzzy

5.1.1 Artificial (Computational) Neural Network

The HCI designer can use neural networks to implement many aspects of the FIS. (Note: if the reader is unfamiliar with neural networks, an excellent overview by Dr. Leslie Smith can be found here.) For example, a neural network can be used to implement:

1. Membership function

2. Fuzzy rules

3. Defuzification

5.1.2 Adaptive neuro-fuzzy inference system (ANFIS)

As mentioned in section 4.2, a the parameters of a Sugeno FIS can be optimized automatically using a computer algorithm. One such algorithm, the adaptive neuro-fuzzy inference system (ANFIS), adapts the parameters of the FIS using neural networks. The original paper on ANFIS can be found in "ANFIS: Adaptive Network Based Fuzzy Inference Systems," by J - S. R. Jang in IEEE Transactions on Systems, Man, and Cybernetics, Vol. 23, No. 3, pp. 665-685, May 1993.

To use ANFIS, the HCI designer needs to perform the following steps:

1. Design a Sugeno FIS approriate for the classification problem.

2. Hand optimize the FIS given actual input classification data.

3. Set up training and testing matrices. The training and testing matrices will be composed of inputs and the desired classification corresponding to those inputs.

4. Run the ANFIS algorithm on the training data.

5. Test the results using the testing data.

Previous Page