News or social media? Socio-economic divide of mobile service consumption

Abstract

Reliable and timely information on socio-economic status and divides is critical to social and economic research and policing. Novel data sources from mobile communication platforms have enabled new cost-effective approaches and models to investigate social disparity, but their lack of interpretability, accuracy or scale has limited their relevance to date. We investigate the divide in digital mobile service usage with a large dataset of 3.7 billion time-stamped and geo-referenced mobile traffic records in a major European country, and find profound geographical unevenness in mobile service usage—especially on news, e-mail, social media consumption and audio/video streaming. We relate such diversity with income, educational attainment and inequality, and reveal how low-income or low-education areas are more likely to engage in video streaming or social media and less in news consumption, information searching, e-mail or audio streaming. The digital usage gap is so large that we can accurately infer the socio-economic status of a small area or even its Gini coefficient only from aggregated data traffic. Our results make the case for an inexpensive, privacy-preserving, real-time and scalable way to understand the digital usage divide and, in turn, poverty, unemployment or economic growth in our societies through mobile phone data.

Awakening City: Traces of the Circadian Rhythm within the Mobile Phone Network Data

“…In a previous work, we also demonstrated differences in mobility customs between socioeconomic classes [27] in the case of Budapest. Ucar et al revealed the socioeconomic gap by mobile service consumption [33]. Vilella et al found that education and age play a role in news media consumption patterns in Chile, using a dataset that provides information about the visited websites [34].…”

Towards a new paradigm for segregation measurement in an age of big data

“…Another possible solution is to infer users' socioeconomic attributes based on their behaviour and mobility patterns. For instance, a study using mobile-app data found that income status and educational attainment are highly related to mobile service (e.g., news, e-mail, social media consumption and video) usage, although the evidence was yielded at the aggregate level (Ucar et al, 2021). Through one-week GPS records of over 400 volunteers, a case study demonstrated the strong associations between demographic attributes (migrant status, marital status, education) and trajectory patterns (Wu et al, 2019).…”

“…Theoretically, these virtual interactions are not constrained by geographic distances. Moreover, different social groups may have distinct preferences for certain online platforms and services for access to information and social activities (Ucar et al, 2021). Therefore, it rises a hot concern about whether people experience social inequalities and processes similar to that in physical spaces (Li & Wang, 2014).…”

AppShot: A Conditional Deep Generative Model for Synthesizing Service-Level Mobile Traffic Snapshots at City Scale

“…In Figure 11, we label the dimensions of input/output at each layer using the format of [channels, size x , size y ]. For example, the c [27,12,12] on the input side means the context input is a multi-channel image with 12 × 12 dimensions and 27 channels; each channel here represents a particular condition (i.e., contextual attribute). To reduce over-fitting to the conditions, we add a channel-wise dropout layer to the input conditions (with a dropout rate of 0.02).…”

“…Equally, such data is crucial in planning and configuration of edge computing and storage infrastructure to optimize latency for certain services and benefit others by computation offloading to the edge cloud [5], [6]. Furthermore, service-level traffic data is key to other service oriented studies within networking, including the design of traffic classification techniques [7] and energy efficiency optimization [8], and beyond, such as for data plan analysis [9], [10] or to reveal links between apps consumption and urbanization levels or socioeconomic status [11], [12].…”