This year we celebrate the 100th Anniversary from the birthday of Acad. Olexiy Gryhorovych Ivakhnenko (30.03.1913-16.10.2007),
the founder of the scientific school of inductive modelling and creator of the world-wide known GMDH method.
Group
Method of Data Handling* was applied in a great variety of areas for data mining and knowledge discovery, forecasting and systems modeling, optimization and pattern recognition. Inductive GMDH algorithms give possibility to find automatically interrelations in data, to select optimal structure of model or network and to increase the accuracy of existing algorithms.
This original self-organizing approach is substantially different from deductive methods used commonly for modeling. It has inductive nature - it finds the best solution by sorting-out of possible variants.
By sorting of different solutions GMDH networks aims to minimize the influence of the author on the results of modeling. Computer itself finds the structure of the model and the laws which act in the system.
Group Method of Data Handling is a set of several algorithms for different problems solution. It consists of parametric, clusterization, analogues complexing, rebinarization and probability algorithms. This inductive approach is based on sorting-out of gradually complicated models and selection of the best solution by minimum of external criterion characteristic. Not only polynomials but also non-linear, probabilistic functions or clusterizations are used as basic models.
GMDH approach can be useful because:
- The optimal complexity of model structure is found, adequate to level of noise in data sample. For real problems solution with noised or short data, simplified forecasting models are more accurate.
- The number of layers and neurons in hidden layers, model structure and other optimal NN parameters are determined automatically.
- It guarantees that the most accurate or unbiased models will be found - method doesn't miss the best solution during sorting of all variants (in given class of functions).
- As input variables are used any non-linear functions or features, which can influence the output variable.
- It automatically finds interpretable relationships in data and selects effective input variables.
- GMDH sorting algorithms are rather simple for programming.
- TMNN neural nets are used to increase the accuracy of another modelling algorithms.
- Method uses information directly from data sample and minimizes influence of apriori author assumptions about results of modeling.
- Approach gives possibility to find unbiased physical model of object (law or clusterization) - one and the same for future samples.
It was implemented in many commercial software products.
|
