Artificial intelligence has been applied in many scientific fields the last years with the development of new neural network technologies and machine learning techniques. In this research, artificial neural networks are implemented for developing prediction models in order to forecast unemployment. A Feedforward Neural Network architecture was applied, since it is considered as the most suitable in times series predictions. The best artificial neural network forecasting model was evaluated by testing different network topologies regarding the number of the neurons, the number of the hidden layers, and also the nature of the transfer functions in the hidden layers. Several socioeconomic factors were investigated in order to be taken into consideration so as to construct the optimal neural network based forecasting model. The results have shown a very good prediction accuracy regarding the unemployment. The proposed methodology can be very helpful to the authorities in adopting proactive measures for preventing further increase of unemployment which would cause a negative impact on the society.
CITATION STYLE
Kouziokas, G. N. (2019). Unemployment Prediction in UK by Using a Feedforward Multilayer Perceptron. In Springer Proceedings in Business and Economics (pp. 65–74). Springer Science and Business Media B.V. https://doi.org/10.1007/978-3-319-95666-4_5
Mendeley helps you to discover research relevant for your work.