Yuki Akiyama, an associate professor at the Faculty of Architecture and Urban Design at Tokyo City University, has developed a map that predicts the rate of vacant houses in local governments nationwide.Based on the census of the Ministry of Internal Affairs and Communications and the housing and land statistics survey, it was possible to make an estimate including areas that were not covered by the survey.
According to Tokyo City University, Associate Professor Akiyama used AI (artificial intelligence) to analyze population trends and changes in the number of vacant houses from the past to the present in each municipality nationwide based on the Ministry of Internal Affairs and Communications' national census and housing and land statistics survey. We have developed a technology to learn and predict the vacant house rate in each municipality in 2018, 2023, and 2028.
It is possible to make highly accurate predictions with an average error of 1.96%, and it is possible to calculate the vacant house rate for the entire municipality, including areas where field surveys have not been conducted.Associate Professor Akiyama has published an empty house map that can be viewed on the web.It is said to be the first vacant house map that covers all municipalities nationwide.
In response to the rapid increase in vacant homes nationwide, the government is asking local governments nationwide to grasp the distribution of vacant homes, but an efficient method for surveying vacant homes has yet to be established.In some municipalities, even the current number of vacant houses is unknown because they are not covered by government statistics.
In the future, Associate Professor Akiyama plans to calculate the vacant house rate by region from the current municipality level by using government statistics micro data.By using this data, it is expected that more efficient countermeasures against vacant houses will become possible.