CHARACTERISTICS OF HIGHLY RATED RESTAURANTS CONCENTRATION AREAS IN THE URBAN CENTERS OF SAPPORO, SENDAI, HIROSHIMA, AND FUKUOKA

In this study, highly rated restaurants concentration areas were extracted by social media and categorized based on restaurant characteristic indicators. Then, machine learning models were used to characterize the environment by area typology and to evaluate the potential for non-concentrated areas of highly rated restaurants. As a result, it was found that the characteristics of restaurants in highly rated restaurants concentration areas differ for each surrounding environment. Furthermore, by using a model with high prediction accuracy, it showed the possibility of predicting the characteristics of restaurants to be located based on the characteristics of the surrounding environment.