GMDH Shell is an advanced but easy to use tool for data mining, predictive analytics, time series analysis, forecasting and knowledge discovery. The software combines well proven machine learning technology and extended capabilities for effective use of multi-core, multiprocessor and clustered computers.
GMDH Shell makes processing of data much easier in many aspects. It is able to automatically detect usable data inside a file, transform data according to a problem type, drop irrelevant inputs and, finally, construct a set of predictive models at the base of optimal complexity detection and self-organization principles.
Notable capabilities of the software environment:
- Time series forecasting, classification and continuous value prediction.
- GMDH-type neural networks.
- Linear and non-linear multiparametric GMDH models.
- Estimation of variables' importance. Feature selection.
- Processing of CSV and XLS files including sets of files.
- Handling of missing values. Text variable detection.
- Encoding and binary decomposition of text variables.
- One-click result recalculation for dynamically updated data files.
- Background execution mode via command line.
- Support for muti-core processors.
- Support for HPC systems (HPC edition only).
![]() |
![]() |
![]() |





