A Study on Employment-centric Industries in Incheon Using Machine Learning Method

The industrial structure in Incheon has a very significant difference compared with the change in the industrial structure of the whole country. Thus, reflecting those differences, the unemployment rate in Incheon has always been higher than the national average. The purpose of this study is to identify the employment - oriented industries in Incheon area using machine learning techniques. First, the relationship between employment ratio and employment by industry was examined through regression analysis. Second, we examined the characteristics of industrial areas with high employment rate through decision tree analysis. Third, a random forest model was used to predict regional employment changes. Through such a machine learning model, the Incheon area still has a base of traditional manufacturing, but it is characterized by the expansion of the service industry such as wholesale and retail business. Machine learning methods have multiple tools analyzing and reporting complicated non-linear relationship in data effectively. In addition, a random forest model was used to predict employment changes due to industrial structure changes. This machine learning model shows that beverage manufacturing industry has the highest employment rate in Incheon and expansion of service industry such as wholesale and retail business is prominent.