Home

APS
Downloads
Buy APS

Support
Register
Contact us

Knowledge Base of APS

Subscribe to the GEPList
 
Visit GEP

 

Introduction

 
 

Automatic Problem Solver is an extremely flexible modeling tool designed for Function Finding, Classification, and Time Series Prediction.

There are four different editions of APS: Standard, Advanced, Professional, and Academic. The Academic Edition has all the features of the Professional Edition and can only be used for academic purposes. See a feature comparison table of the four editions here.

APS is very easy to use and as soon as you learn how to load your data, you will be able to generate a model immediately as APS comes with a series of templates that only require a mouse click for you to design a model.

With APS you can fully understand and analyze the evolved models as APS automatically translates the models evolved in its native Karva code into a wide set of programming languages (C, C++, C#, Visual Basic, VB.Net, Java, Java Script, and Fortran) through the use of built-in grammars. Furthermore, APS 3.0 also allows you to create your own grammars for translating the native Karva code, and, therefore, with APS 3.0 you can automatically translate all the evolved models into virtually any programming language through the use of User Defined Grammars.

You can then paste the evolved models into your IDE or code files and use them to create highly sophisticated software applications such as software for predicting unknown behavior, classifying previously unclassified samples, forecasting future events based on past ones, and so on.

Not withstanding, with APS 3.0 you don’t have to leave the APS environment for making predictions about the future or predicting unknown behavior with the evolved models, as APS 3.0 comes equipped with a scoring engine that allows you to immediately apply the evolved models to an unlimited number of records. Indeed, as soon as a model is created by APS, it can be immediately used for making predictions or extrapolations exactly as it is, with the advantage that one doesn’t have to know how to program to do that.

The modeling algorithms of APS are based on Gene Expression Programming (GEP), an extremely fast and powerful learning algorithm. APS can be used without understanding Gene Expression Programming, but if you do you will be able to explore more thoroughly all of its features. In this Help file, you’ll find all the necessary information, both theoretical and practical, for exploring all APS features successfully. For complementary information, though, we recommend the seminal GEP paper published in the journal Complex Systems:

Ferreira, C., 2001. Gene Expression Programming: A New Adaptive Algorithm for Solving Problems, Complex Systems, 13 (2): 87-129,

which is available in pdf format at the Gene Expression Programming website and the book:

Ferreira, C., 2002. Gene Expression Programming: Mathematical Modeling by an Artificial Intelligence, Angra do Heroismo, Portugal,

also available at the Gene Expression Programming website.

Home | Contents | Previous | Next