Personnel classification for human resource allocation - A neural network approach

While personnel classification based on personality data is considered as the most important decision-making problem in human resource allocation, the nature of subjective and incomplete evaluation of persons to be classified prevents any quantitative method from being useful. We suggest the use of unsupervised Competitive Learning neural networks with the Input Feature Scaling algorithm to overcome the obstacles that conventional analyses face. Simulations using real data prove the potential of the approach. The performance of the approach is comparable to that of the probabilistic-error-minimizing Bayes classifier.