当前位置: X-MOL 学术Big Data › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Mining the Thin Air—for Understanding of Urban Society
Big Data ( IF 4.6 ) Pub Date : 2019-12-01 , DOI: 10.1089/big.2019.0026
Ron Bekkerman 1, 2 , Adi Zmirli 2 , Scott Kirkpatrick 3
Affiliation  

We explore the potential of crowd-sourced information on human mobility and activities in an urban population drawn from a significant fraction of smartphones in the Los Angeles basin during February-May 2015. The raw dataset was collected by WeFi, a smartphone app provider. The dataset is noisy, irregular, and lean; however, it is large scale (over a billion events), cheap to collect, and arguably unbiased. We employ the state-of-the-art Big Data techniques to turn this structurally thin dataset into semantically rich insights on commuting, overworking, recreational traveling, shopping, and fast food consumption of the Greater LA population. For example, we reveal that Greater LA residents commute substantially longer than what is reported in the US census data. Also, we show that younger individuals dine at McDonald's significantly more than the older population does. Our results have implications for public health, inequality, urban traffic, and other research areas in social sciences. The large number of phones participating in our "crowd" makes it possible to obtain those results without the risk of compromising individual privacy.

中文翻译:

挖掘稀薄的空气-了解城市社会

我们探索了在2015年2月至5月期间从洛杉矶​​流域的大部分智能手机中提取的有关城市人口中人类活动和活动的众包信息的潜力。原始数据集由智能手机应用提供商WeFi收集。数据集嘈杂,不规则且稀疏;但是,它规模宏大(超过十亿个事件),收集起来便宜,并且可以说是公正的。我们采用最先进的大数据技术,将这种结构上薄的数据集转化为大洛杉矶地区通勤,劳累,休闲旅行,购物和快餐消费的语义丰富的见解。例如,我们发现大洛杉矶居民的通勤时间比美国人口普查数据中记录的时间长得多。此外,我们显示,年轻人在麦当劳用餐 远远超过老年人口。我们的结果对公共卫生,不平等,城市交通以及社会科学的其他研究领域具有影响。参加我们“人群”的大量电话使获得这些结果成为可能,而没有损害个人隐私的风险。
更新日期:2019-12-01
down
wechat
bug