Application of Artificial Neural Network model to an Analysis of the Factors Affecting the Intention of the Vulnerable Class to move to Hangbok Housing in Incheon

Abstract

This study aims to examine the impact factors of the intention to live in Hangbok housing, focusing on the vulnerable in Incheon. As analysis method, Artificial neural network(ANN) and binomial logit model are used. Comparative analysis between the methods and synthesis were conducted. The main findings are as follows. First, based on the AUROC and prediction accuracy result, the statistical power of ANN model is better than those of the logit model. Second, important factors that affect the intention to move to Hangbok housing are newlyweds, monthly income, housing benefit recognition, and tenure type. Third, the factors of number of households, tenure type, moving plans, newlyweds, and housing support program have a significant and positive effect on the intention to move to Hangbok housing, while factors of age, social housing living, monthly income, debt status, housing benefit recognition have a negative influence on the intention. Next, it was found that newly-weds who are relatively young and need for housing support programs have a higher willingness to move in the public rental housing, compared to other household types. In addition, it can be seen that households who are currently living in the public rental housing in Incheon do not prefer to change the type of housing to a Hangbok housing.