The Interrelation Miner Methodology is a technology for multivariate statistical data analysis and outcome prediction (data mining) of either complex and/or multidimensional data sets, or of very small datasets based on statistical fuzzy set theory. This tool can be applied to any complicated data set with many samples or variables (e.g., micro array data), and provides results for small or incomplete data tables by using very simple decision rules with nonlinear data models. The Interrelation Miner has analyzed gas chromatography/mass spectrometry data and has provided up to 20% better prediction performance than compared approaches. Other possible applications include: microarrays, protein arrays, environmental data sets, food quality control, and quality control in general.
Time for multivariate statistical development is minimized as the technology reveals significant simple structures within complex data. It maximizes the quality of multivariate statistical predictions because it improves accuracy of predictions based on complex data by up to 20%. The Interrelation Miner can be used for machine learning and permanent knowledge acquisition.
Get the complete report on this technology at: