Approximation using the IBHM method

Description

Implementation of an incremental model construction method called IBHM which stands for Incrementally Built Heterogeneous Model. The method is designed for solving real-number approximation problems in a highly automated fashion and should be considered one of black-box modelling approaches such as Artificial Neural Networks or Support Vector Machines.

Details

The package may be used to produce small, robust computational models using real valued empirical data. The most important functions are TrainIBHM which constructs an IBHM model and ConfigureIBHM which sets the methods parameters.

Author(s)

Pawel Zawistowski

Maintainer: Pawel Zawistowski <dratewka@gmail.com>

References

Zawistowski, P. and Arabas, J.: "Benchmarking IBHM method using NN3 competition dataset." In Proc. 6th int. conf. on Hybrid artificial intelligent systems - Vol. 1, HAIS'11, pp 263–270, 2011. Springer-Verlag.

Results