Application of Association Rules and an Artificial Neural Network to Predict the Urban Development of Regional Revitalization

Abstract

With the decrease in Taiwan’s total population and its dense concentration in metropolises, an imbalance between urban and rural development has emerged. Meanwhile, regional revitalization (RR) and the regeneration of local industries have been considered sustainable methods for regional development. However, a complete evaluation system is still needed to determine suitable directions and models for RR projects. Based on successful RR projects and indicators of the RR database in Taiwan, this study categorizes 55 projects into four types of RR. The association rules method is adopted to explore the relationship between the indicators and the RR types. Subsequently, an artificial neural network (ANN) model is developed to predict the future adoption of revitalization development. The results confirm that the characteristics of different regions are closely related to RR development. On the other hand, the findings reveal that the ANN model can verify the prediction accuracy of future RR models. This study also proposes a double-diamond framework to evaluate and improve the prediction of RR development, which is expected to help the government develop plans and devise strategies in different regions in the future.