BrainCom can be used in almost all research fields such as nonlinear regression, forecasting, curve fit, pattern recognition, classification, decision making, problem optimization, time series analysis.

Its user friendly environment makes it very easy to handle even by a novice. BrainCom can handle unlimited number of input vectors, inputs, outputs and hidden layer nodes.

You can create your training files using your favorite spreadsheet and import them to BrainCom as csv files. All other files created are of the same format.

An ActiveX control is included to use it in your own programs to make calculations in real time.

Here are some key features of "BrainCom":

￭ Accepting csv format files from spreadsheets like Microsoft's Excel.

￭ Networks with one or two hidden layer topology

￭ Adjustment of training parameters such as momentum and learning rate.

￭ Set the size of hidden layers.

￭ Automatic data normalization

￭ Two kinds of activation function: Logistic and Hyperbolic tangent

￭ Fast training algorithm

￭ Charts of total error progress and error distribution

￭ A status panel showing the current status of the network

￭ You can set command line parameters. You can write your own programs to create the data and call BrainCom with command line parameters to train networks with data you've just created. After you've trained a network you can call again BrainCom to make predictions with new input data.

Limitations:

￭ 30 days trial