Measuring Chinese cities’ economic development with mobile application usage

With the rise of smart phones, mobile applications have been widely used in daily life. However, the relationship between individuals’ mobile application usage and cities’ economic development has yet to be investigated. To study this question, this work utilizes a dataset containing users’ history of mobile application usage records (MAURs) and investigates how MAURs are related to Chinese cities’ economic development. Our analysis shows the cities’ GDP and number of MAURs are highly correlated, and at the individual level, people in wealthier cities (higher GDP per capita) tend to have more active mobile application usage (MAURs per capita). The results also demonstrate the relevance between cities’ GDP and MAURs varies significantly among different demographic groups, with male users’ relevance consistently higher than female users’ and working-age people’s relevance higher than other age groups. A boosted tree regression model is then applied to predict cities’ GDP with MAURs and proves to achieve high goodness-of-fit (over 0.8 R-square) and good prediction accuracy, especially for the economically developed and populous regions in China. To the best of our knowledge, this is the first time that the relationship between MAURs and cities’ economic development is revealed, which contributes to novel knowledge discovery for regionalization and urban development.