Deep Learning: Practical Neural Networks with Java

Input selection

One of the key tasks in designing a neural network application is to select appropriate inputs. For the unsupervised case, one wishes to use only relevant variables on which the neural network will find the patterns. And for the supervised case, there is a need to map the outputs to the inputs, so one needs to choose only the input variables which somewhat have influence on the output.

Data correlation

One strategy that helps in selecting good inputs in the supervised case is the correlation between data series, which is implemented in Chapter 5, Forecasting Weather. A correlation between data series is a measure of how one data sequence reacts or influences the other. Suppose we have one dataset containing a number of data series, ...