NeuroSolutions 4.3 is now available

11 March, 2004

NeuroSolutions 4.3 is now available - powerful and flexible neural network modeling software.

NeuroSolutions 4.3 is now available! This powerful and flexible neural network modeling software is the perfect tool for solving your data modeling problems.

Neural networks and genetic algorithms are exciting technologies in the field of artificial intelligence which mimic the learning process of the human brain in order to extract patterns from historical data. These technologies have steadily changed the way we solve "real-world" problems in science, engineering and economics. The development tool of choice among neural network researchers and application developers is NeuroSolutions.

Some of the most recent additions to NeuroSolutions:

  • Neuro-Fuzzy - The coactive neuro-fuzzy inference system (CANFIS) model integrates fuzzy inputs with a neural network to quickly solve poorly defined problems. Fuzzy inference systems are also valuable as they combine the explanatory nature of rules (membership functions) with the power of neural networks.


  • Support Vector Machine - The Support Vector Machine (SVM) model maps inputs to a high-dimensional feature space, and then optimally separates data into their respective classes by isolating those inputs that fall close to the data boundaries. They are especially effective in separating sets of data that share complex boundaries.


  • Conjugate Gradient - Conjugate gradient learning is a second-order training method that provides an excellent trade-off between complexity and performance. Typically it trains faster and better (lower MSE) than standard backpropagation. In addition, it is completely parameterless -- no learning rates or momentum terms to adjust.


  • Teacher Forcing / Iterative Prediction - There are some time-series prediction problems that are best modeled using a method called teacher forcing. This specialized training algorithm feeds the predicted output back into the input in order to improve the accuracy of multi-step prediction. The predicted output of networks trained with teacher forcing is then obtained using iterative prediction.

"I have recently purchased a copy of NeuroSolutions 4 and am very happy with the software. It is amazing how many features are available within the network."
- Albrecht Stoecklein (MSc), Building Research Association of New Zealand

NeuroSolutions' icon-based graphical user interface provides the most powerful and flexible development environment available on the market today. Its intuitive wizards and optional Excel interface make it quick and easy to build and train a neural network to solve your problem. From there, you can easily deploy your neural network solution to a custom application. Download a free evaluation copy of NeuroSolutions now to learn more about how to put this remarkable technology to work for you.

Neural network problems can generally be categorized as one of four types. NeuroSolutions is one of the few products on the market that is able to handle all of these types of problems.

Classification
Classification problems are those where the goal is to label each input pattern as belonging to a certain class. A simple example of a classification problem is one where the goal of the neural network is to label each person as male or female (the two classes) based on their height and weight. The input into the neural network would be the height and weight measurements and the desired output would be their gender.

Function Approximation
Function Approximation problems are those where the goal is to determine a numeric value given a set of inputs. This is similar to classification problems except that the output is numeric. An example is to determine the wind chill factor (the desired output) given the temperature, humidity, and wind speed (the inputs). These problems are called function approximation because the neural network will try to approximate the functional relationship between the input and desired output.

Prediction
Prediction problems are those where the goal is to determine a future output given a set of inputs and the past history of the inputs. The main difference between prediction problems and the others is that prediction problems use the current input and previous inputs (the temporal history of the input) to determine a future output value. A typical example is to use the temporal history of a stock closing price as input (e.g., today’s and three previous day’s prices) to try to predict tomorrow’s closing price.

Clustering
Clustering problems are those where you want the neural network to extract information from only the input data. For instance, you have survey data from various people. You would like to cluster (partition) the people into groups with similar buying habits in order to better target your marketing efforts. The fundamental difference between the clustering problem and the others is that there is no desired output.

Free Worldwide Delivery for Online Orders

Order online from Hearne Scientific Software and we will ship NeuroSolutions to you for FREE.

For more information regarding NeuroSolutions please contact Hearne Scientific Software.

Rate this page
Comment