ANALYSE D’UNE BASE DE DONNEE DES SEMI-CONDUCTEURS CHALCOPYRITE EN 2 ET 3 DIMENSIONS

The analysis of the data is currently, in the statisticians, the object of a true phenomenon of mode, characterized at the same time by the passion and the rejection. To determine the characteristics of a whole of data is obviously not an end in itself, but this lapse of memory is unfortunate because the type of data available always influences the direction of research. The factorial analysis in normalized principal components (PCA), is particularly adapted to the reduction of tables of measurements, where one generally works on a matrix of correlation between characters. In this article we will report our results of analysis into 2 and 3 dimensions of the molecular weight and energy of plasmon of semiconductor chalcopyrites by using what one call the datamining materials. All information was presented in the tables envisaged by an ACP method of the semiconductors etudies. L' analyzes factorial substitutes for primitive measurements of new variables or factors (which are linear combinations of the primitive variables) with the help of which it is thus possible to optimize the visualization of the data in a restricted number of figures. Large at the modo we present, the results related to the development of new approaches of Datamining, Analysis in principal components (ACP), of the data-processing techniques which were applied to semiconductors and combined with metals to predict new materials or ultra stable coatings.