Data mining: An application to the semiconductor industry

The development project consists of the study and use of data mining techniques to analyze a dataset of high dimensionality and large volume generated during the manufacture of memory modules in a semiconductor industry. In this paper, we propose the use of a neural network self-organizing map as a technique for knowledge extraction, we discuss the analysis phase and preparing the database for the mining process and presents the following steps research. It is expected that the results at the end of the project to identify relationships and / or patterns that may help in predicting possible factors causing failures during the production process, thus contributing to improving the quality of industrial processes, particularly in semiconductor industry.