Big data dig up China's "ghost cities"
                 English.news.cn | 2015-11-06 18:07:50 | Editor: huaxia

Picture shows Rushan, a seaside town in Weihai city of eastern China's Shandong Province, which is believed to be the most problematic ghost city with its large numbers of unclaimed ocean-view houses. (Web Photo)

BEIJING, Nov. 6 (Xinhua) -- While the Chinese government has yet to reveal any official data on the vacancy rates of the country's real estate, a research team from Internet search engine company Baidu Inc. believe they have got the answer using big data technology.

Researchers at Baidu's big data lab and Peking University used positioning data collected from 770 million Chinese Internet users to draw "the most accurate picture" of China's "ghost cities", or vacant housing areas.

The results of their study, published recently on the website of Cornell University Library http://arxiv.org, present the spatial distribution of the vacant housing areas in China and classifies cities with a large vacant housing area as cities or tourism sites.

Amid the real estate boom in the past decade, many new housing districts are built excessively across China, but they far exceed the actual demand in some cities, according to their analysis.

To understand the human dynamic in "ghost cities", researchers select one city and one tourism site as cases to analyze the features of human dynamics.

Researchers captured billions of data information containing user IDs, latitudes, longitudes, and duration of stay everyday from 9 a.m. to 6 p.m. between September 2014 and April 2015.

The results showed the underpopulated areas are mainly located in tourist attractions, third-tire cities, new towns and urban centers. Rushan, a seaside town in Weihai city of eastern China's Shandong Province, was believed to be the most problematic one with its large numbers of unclaimed ocean-view houses.

Most "ghost cities" with a large vacant housing area are mostly second-tier and third-tier cities, with provinces in the east having more proportion of cities with vacant housing areas.

According to local newspaper Shandong Business Daily, Yintan, a real estate company, has developed more than 18 million square meters of land in Rushan during the past 20 years. But 85 percent of the houses are vacant.

In total, twenty districts were listed as highly-severe "ghost cities". Some were previously unknown, like Dongying, an overbuilt oil boom town. Kangbashi New Area, one that was most covered by the media, is fairly well populated now and off the list.

The features of national spatial scale, long temporal scale and high precision of Baidu big data make the study of "ghost cities" representative and reliable, they said.

Instead of just counting the number of homes with light at night in certain residential areas as the indicator of "ghost city", Baidu big data can count the population precisely, in real time, and in national scale.

Still, they admitted that the big data technology is limited as it can't represent the real demography of a city because not all people are Baidu users.

The results discovered the specific location of vacant housing areas, which can help government make smarter and more reasonable decisions.

They also distinguish the tourism sites and cities.

Back to Top Close
Xinhuanet

Big data dig up China's "ghost cities"

English.news.cn 2015-11-06 18:07:50

Picture shows Rushan, a seaside town in Weihai city of eastern China's Shandong Province, which is believed to be the most problematic ghost city with its large numbers of unclaimed ocean-view houses. (Web Photo)

BEIJING, Nov. 6 (Xinhua) -- While the Chinese government has yet to reveal any official data on the vacancy rates of the country's real estate, a research team from Internet search engine company Baidu Inc. believe they have got the answer using big data technology.

Researchers at Baidu's big data lab and Peking University used positioning data collected from 770 million Chinese Internet users to draw "the most accurate picture" of China's "ghost cities", or vacant housing areas.

The results of their study, published recently on the website of Cornell University Library http://arxiv.org, present the spatial distribution of the vacant housing areas in China and classifies cities with a large vacant housing area as cities or tourism sites.

Amid the real estate boom in the past decade, many new housing districts are built excessively across China, but they far exceed the actual demand in some cities, according to their analysis.

To understand the human dynamic in "ghost cities", researchers select one city and one tourism site as cases to analyze the features of human dynamics.

Researchers captured billions of data information containing user IDs, latitudes, longitudes, and duration of stay everyday from 9 a.m. to 6 p.m. between September 2014 and April 2015.

The results showed the underpopulated areas are mainly located in tourist attractions, third-tire cities, new towns and urban centers. Rushan, a seaside town in Weihai city of eastern China's Shandong Province, was believed to be the most problematic one with its large numbers of unclaimed ocean-view houses.

Most "ghost cities" with a large vacant housing area are mostly second-tier and third-tier cities, with provinces in the east having more proportion of cities with vacant housing areas.

According to local newspaper Shandong Business Daily, Yintan, a real estate company, has developed more than 18 million square meters of land in Rushan during the past 20 years. But 85 percent of the houses are vacant.

In total, twenty districts were listed as highly-severe "ghost cities". Some were previously unknown, like Dongying, an overbuilt oil boom town. Kangbashi New Area, one that was most covered by the media, is fairly well populated now and off the list.

The features of national spatial scale, long temporal scale and high precision of Baidu big data make the study of "ghost cities" representative and reliable, they said.

Instead of just counting the number of homes with light at night in certain residential areas as the indicator of "ghost city", Baidu big data can count the population precisely, in real time, and in national scale.

Still, they admitted that the big data technology is limited as it can't represent the real demography of a city because not all people are Baidu users.

The results discovered the specific location of vacant housing areas, which can help government make smarter and more reasonable decisions.

They also distinguish the tourism sites and cities.

[Editor: huaxia ]
010020070750000000000000011103261347911491