6.8 APPLICATION OF ARTIFICIAL NEURAL NETWORKS
1.classification. A neural network can be trained to predict a categorical(i,e., class-label) output variable.
In a mathematical sense, this involves dividing an n-dimensional space into various regions and given a point in the space one should real-world applications of pattern recognition, where each pattern is transformed into a multidimensional point and classified into a certain group, each of which represents a known pattern.
Type of ANN used for this task include feedforward networks (such as MLP with backpropagation leaning), radial basis functions, and probabilistic neural networks.