New in Versions 3.0 and 3.1!
Includes facility to create Self Organizing Maps (SOMs) to identify
clusters and associations in data
Improved performance due to removing code for "thunking" layer
previously required by now obsolete operating systems
Now available for 32- and 64- bit UNIX, and selected Linux platforms
(Command Line only).
Revised and updated Tutorial and documentation.
Better visualization of processes.
Menus for Excel add-in have been re-arranged for better consistency.
New packaging and packaging design.
NeuralWorks Predict is an integrated, state-of-the-art tool for rapidly
creating and deploying prediction and classification applications.
Predict combines neural network technology with genetic algorithms,
statistics, and fuzzy logic to automatically find optimal or
near-optimal solutions for a wide range of problems. Predict
incorporates years of modeling and analysis experience gained from
working with customers faced with a wide variety of analysis and
Predict requires no prior knowledge of neural networks. With only
minimal user involvement it addresses all the issues associated with
building robust models from available empirical data. Predict analyzes
input data to identify appropriate transforms, partitions the input data
into training and test sets, selects relevant input variables, and then
constructs, trains, and optimizes a neural network tailored to the
problem. For advanced users, Predict also offers direct access to all
key training and network parameters.
In Microsoft® Windows environments NeuralWorks Predict can be run either
as an add-in for Microsoft Excel to take advantage of Excel's rich data
handling and graphing capabilities, or as a command line program that
offers powerful batch mode processing. In UNIX and Linux environments,
NeuralWorks Predict runs as a command line program.
With 1 Year TAP Subscription
|Sun Sparc/Solaris - 32 bit
|Sun Sparc/Solaris - 64 bit
|Silicon Graphics/Irix - 32 bit
|Silicon Graphics/Irix - 64 bit
All Prices in US dollars. Prices effective 1 July 2001. Prices subject
to change without notice.
All products are normally distributed on CD-ROM. Please contact
NeuralWare at email@example.com if you would like another
Academic pricing is available. Please contact NeuralWare at
firstname.lastname@example.org for eligibility requirements and pricing.
If you are interested in using NeuralWorks Predict on a platform other
than those listed here, or if you would like information on quantity
pricing, please contact NeuralWare at email@example.com.
NeuralWorks Predict Features
Easy To Use
In the Microsoft Windows environment, Predict runs as an Add-in with Microsoft Excel. This means both a familiar interface as well as access to a wide variety of ancillary capabilities that you can employ to put Predict outputs into formats suitable for presentation to or use by a wide variety of audiences.
The basic process of building a model with Predict is completely automated. With the Build Wizard activated, a short series of dialog boxes provide a step-by-step guide. Five small dialogs determine the layout of data within a spreadsheet.
Five more small dialogs determine the complexity of the application and how hard and long Predict should run to produce a solution. Creating and training a model is that easy.
Neural network results are written back into the spreadsheet, where a quick graph makes the results suitable for presentation to a wide variety of audiences.
Three levels of interface accommodate the diverse needs of new users, application engineers, and neural network engineers. The Build Wizard guides model building with a sequence of high level dialogs which do not require any detailed understanding of the model components. Advanced and Expert modes provide increasing levels of access to all the details of the internal algorithms.
The complete system has five main components.
- The Train/Test Selection component picks out training and test sets for
model building. It tries to do this in such a way that the test set is statistically close to the training set. It also allows you to hold back a portion of you
r data for independent validation of your model.
- The Data Analysis and Transformation component automatically analyzes data and transforms it into forms suitable for Neural Networks. The types of function that this component performs is to expand categorical data into numeric data, to shape numeric data to get rid of skewness and other undesirable characteristics, to deal with outliers in the data, and to screen out data that contain no information.
- The Input Variable Selection component uses a genetic algorithm to search for synergistic sets of input variables which are good predictors of the output. Because of the evolutionary nature of the Input Variable algorithm, different initializations of the algorithm will yield different variable sets. You can use this to your advantage to build several models based on different variable sets and combine the outputs of those models. Each model can be thought of as an expert which uses a different set of criteria (the selected variables) to make its decision. There is also option to do a pre-selection of variables using a Cascaded genetic algorithm approach. This method gives more consistent variable sets by pruning out variables which are consistently rejected by different invocations of the genetic algorithm.
- The Neural Net component of Predict supports two proprietary non-linear feed-forward constructive algorithms. One of algorithms is based on a non-linear Kalman filter learning rule and is designed for noisy regression problems. The other is a general-purpose algorithm which is based on an Adaptive Gradient learning rule. Multiple networks may be trained for optimal results. Regularization mechanisms provide good generalization. Eleven different evaluation functions are provided for evaluating test performance during training. Users can trade off speed of learning for comprehensive solutions.
- The Flash Code component converts the completed model into C, FORTRAN, or Visual Basic code.
Predict automatically performs all the actions necessary to build a
prediction or classification model. A genetic algorithm rapidly builds
and evaluates mini-networks to identify not only which domain inputs are
significant, but also the type of transform function that ultimately
produces the best network. Then the final neural network is constructed,
trained and, and tested. In many situations the resulting network can
be deployed immediately.
Predict incorporates 6 basic transform types and two additional
miscellaneous transforms. The basic transform types are:
- Enumerated Integer
- Enumerated String
- Fuzzy; and
The first miscellaneous transform applies to either of the Enumerated
transform types. If there are two or more categories that rarely occur,
they are combined into a single category labeled Other. The result of
the transform is Tmax if the input field doesn't match any of the
categories of the corresponding Enumerated set; otherwise the transform
result is Tmin.
The second miscellaneous transform applies to missing or invalid numeric
data. If an input field has no data value, or is not a valid numeric
value, the result of this transform is Tmax. Otherwise the result of
this transform is Tmin.
Continuous transforms that are available in Predict include:
- Linear (the identity transform)
- Log (the natural logarithm function)
- LogLog (the logarithm of the logarithm function)
- Exp (the exponential function)
- ExpExp (the exponential of the exponential function)
- Pwr2 (the square function)
- Pwr4 (the fourth power function)
- Rt2 (the square root function)
- Rt4 (the fourth root function)
- Inv (the inverse function - 1/x)
- InvPwr2 (the inverse of the square function )
- InvPwr4 (the inverse of the fourth power function )
- InvRt2 (the inverse of the square root function)
- InvRt4 (the inverse of the fourth root function)
- Tanh (the hyperbolic tangent function)
- LnX/(1-X) (the natural logarithm of (x/(1-x))
Logical transforms that are available in Predict include:
- Reverse Logical
These transforms convert continuous data into logical (two-valued) data,
by comparing the input value to (Inmin + Inmax)/2.
Enumerated Integer and Enumerated String transforms convert ranges of
values into discrete categories. The effect on string (literal)
variables is to produce a 1 of N encoding of the variables.
Fuzzy transforms convert input values to "fuzzy" values between 0 and 1,
based on a user specified transition point identified by left, center,
The Quintile transform uses piece-wise linear transformations to map
input values into 5 bins, such that approximately equal numbers of input
records end up in each bin.