China’s ghost cities – miles of deserted shopping malls, sports stadiums and empty housing developments with zero buyers in sight – lie under the national radar and can go unaccounted for for years.
Now Chinese web services giant, Baidu, has embarked on a new study in which it uses location information from users’ mobile devices, as well as mapping and building data, to identify areas with high volumes of construction with relatively low population densities.
The researchers, in the published findings Ghost Cities: Analysis Based on Positioning Data in China, were also able to discount areas which experienced high levels of tourism which skew the figures in peak seasons.
The web company has a massive 700 million people registered to its online services – a significant proportion of China’s 1.36 billion population. Although the data is not widely representative of Chinese society it is a good representation of urban density. The company argued: “With the ubiquity of smart mobile phones, Baidu users occupy the most proportion of the whole population.”
Tracking its users every day over a six-month period between 2014 and 2015, the scientists used a common clustering algorithm to reveal their home location. This data was then correlated with a further dataset of residential areas to calculate where they lived. Urban density was calculated by the number of people living in each 100 metre by 100 metre area.
The Chinese Ministry of Housing and Urban Rural Development defines an urban region as an area where 1 sq.km houses over 10,000 people. A ghost city, according to Baidu, only houses half of this figure.
The Baidu Big Data team discovered 50 ghost cities by employing this method, although only 20 of these were revealed in the report to avoid potential harm to the real estate market in these areas.
Lead researcher, Guanghua Chi hopes that the study will help the Chinese government to improve city planning processes in the future, as well as supporting individuals thinking about moving to the areas.