Databases Reference
In-Depth Information
Fig. 9.8. Comparison of average classification accuracy of MLP, FSN and OPFSN for
the test sets, X-axis values represent 1: IRIS, 2: WINE, 3: PIMA, and 4: BUPA databases.
Fig. 9.9.
Comparison of performance of different datasets for selecting different size of
features.
9.8. Conclusion
In this chapter, we have discussed the fuzzy swarm net (FSN) model for the
task of classification in data mining. The FSN model expands the given set
of inputs into three categories: low, medium and high. These inputs are fed
to the single layer feed forward artificial neural network. A set of such nets
is being taken to spread in a distributed environment. Swarm intelligence
technique is used to train these nets.
Further we have briefed the concept of Polynomial Neural Network
(PNN) algorithm and its use for handling the classification problems. The
PNN model takes a subset of the features to produce a polynomial called the
