We applied neural networks to quantitative chemical analysis. The input data corresponding to the spectra of X-ray photoelectron spectroscopy (XPS) and Auger electron spectroscopy (AES) were prepared by the sum of two or three peaks with a Gaussian distribution. The neural networks with Kohonen's self-organized feature map and with a back propagation algorithm were used. From the results, we found that it was possible to analyze the patterns without determining the number of peaks, the shapes and so on. Therefore, neural networks are thought to be useful in analyzing XPS and AES data quantitatively.
