Abstract:

Results of testing of the electronic components shipped for the space industry are represented by arrays of data vectors of very high dimensionality, up to hundreds of dimensions. One of the most important problems for increasing the quality of the electronic units is detection of the homogeneous production batches of the electronic devices (components). In this paper, we consider the problem of data mining (fuzzy clustering) with new genetic algorithm with greedy agglomerative heuristic based on the EM-algorithm. Computational experiments with electronic component testing data and classical datasets for clustering problems show that this new algorithm allows obtaining more precise results in comparison with classical EM algorithm and its modifications. Accurate detection of homogeneous production batches reduces the expenses for destructive testing.