Neural network pattern learning for classifying administrators from examples

Values obtained from multiple measurements are used to train a neural network to assess the talents of administrators. A sequence of exercises completed by each administrator in an assessment center provides numeric vectors of twelve skill dimensions. The values in each vector are presumed to have discoverable patterns which make possible the selection of successful administrators or equivalently the removal of the nonsuccessful. The result from the simulated neural network demonstrate the ability of trainable connection-based problem-solvers to build and identify an effective prediction model