Neural Networks Tutorial

1. Application types

Neural networks use information in the form of data to generate knowledge in the form of models.

They are used to discover complex relationships, recognize unknown patterns, predict actual trends and recognize associations from data.

Neural networks have found a wide range of solutions in many different industries. Some of them are the following:

Most of the above applications are belong to one of the following two types:

1.1. Approximation (or function regression)

Approximation can be regarded as the problem of fitting a function from data.

Here the neural network learns from knowledge represented by a training data set consisting of input-target examples. The targets are a specification of what the response to the inputs should be.

In this regard, the basic goal in an approximation problem is to model one or several target variables, conditioned on the input variables.

In approximation applications, the targets are usually continuous variables. Some examples are the following:

A common feature of most data sets is that the data exhibits an underlying systematic aspect, represented by some function, but is corrupted with random noise. The next figure illustrates this fact.

The objective is to produce a model which exhibits good generalization, or in other words, one which makes good predictions for new data.

The best generalization is obtained when the mapping represents the underlying systematic aspects of the data, rather than capturing the specific details (i.e. the noise contribution) of the particular data set.

1.2. Classification (or pattern recognition)

Classification can be stated as the process whereby a received pattern, characterized by a distinct set of features, is assigned to one of a prescribed number of classes. The inputs here include a set of features which characterize a pattern; the targets specify the class that each pattern belongs to.

The basic goal in a classification problem is to model the posterior probabilities of class membership, conditioned on the input variables.

In binary classification, the target variable is usually binary (true or false). Some examples are to:

In multiple classification, the target variable is usually nominal (class_1, class_2 or class_3). Some examples are the following:

As before, the central goal is to design a neural network with good generalization capabilities. That is, a model which is able to classify new data correctly.

⇐ Index Data set ⇒