Estimating Socioeconomic Status via Temporal-Spatial Mobility Analysis - A Case Study of Smart Card Data

Abstract

The notion of socioeconomic status (SES) of a person or family reflects the corresponding entity's social and economic rank in society. Such information may help applications like bank loaning decisions and provide measurable inputs for related studies like social stratification, social welfare and business planning. Traditionally, estimating SES for a large population is performed by national statistical institutes through a large number of household interviews, which is highly expensive and time-consuming. Recently researchers try to estimate SES from data sources like mobile phone call records and online social network platforms, which is much cheaper and faster. Instead of relying on these data about users' cyberspace behaviors, various alternative data sources on real-world users' behavior such as mobility may offer new insights for SES estimation. In this paper, we leverage Smart Card Data (SCD) for public transport systems which records the temporal and spatial mobility behavior of a large population of users. More specifically, we develop S2S, a deep learning based approach for estimating people's SES based on their SCD. Essentially, S2S models two types of SES-related features, namely the temporal-sequential feature and general statistical feature, and leverages deep learning for SES estimation. We evaluate our approach in an actual dataset, Shanghai SCD, which involves millions of users. The proposed model clearly outperforms several state-of-art methods in terms of various evaluation metrics.Index Terms-Socioeconomic Status, smart card, human mobility, data mining, deep learning 1 lendingclub.com, one of the largest peer-to-peer lending platform.

Siamese-Like Convolutional Neural Network for Fine-Grained Income Estimation of Developed Economies

“…To reduce the manpower needed for fine-grained income surveys and to speed up fine-grained income data collection, researchers have used house price as a proxy for income. Previous studies have identified a positive correlation between house price and income [23][24][25][26][27][28][29], whilst house price data are easily accessible and downloadable online in the developed world. However, estimation models that depend on house price as the input and income as the output have yielded a low estimation accuracy.…”

You are how you travel: A multi-task learning framework for Geodemographic inference using transit smart card data

“…Lately, Zhang and Cheng (2018) explored inferring demographics by leveraging a variety of spatial and temporal features extracted from the raw transaction records. Ding, Huang, Zhao and Fu (2019) developed a deep learning model to estimate socioeconomic status using temporal-sequential features and general statistical features generated from SC data. However, the success of these works heavily relied on elaborated feature engineering.…”

User Attribute Inference via Mining User-Generated Data

“…Another important user-generated data type is mobile phone data. However, most of the existing studies only focus on group-level SES inference (at least until the acceptance of our work [25] in 2019). Soto et al explore how to use information derived from the aggregated use of cell phone records to identify the socioeconomic levels of a population [87].…”

“…And in the end, researchers utilize a commonly-used machine learning algorithm, the ridge regression model, to predict the logged income of Facebook users. tweets SES [73] tweets income [58] tweets income [93] tweets education, income [4] tweets occupation, income [40] tweets income [94] tweets education, income [95] tweets education, income [14] tweets family income [61] Facebook Likes income [13] mobile phone metadata personal income [87] mobile phone records SES [29] mobile phone call detail records income [12] mobile phone metadata income [90] mobile phone metadata income [8] cookie income, education level [68] retail transaction records income,education level [96] retail transaction records income, education level [25] smart card transportation records SES [74] WiFi log education, income…”

“…We also collected the volunteers' checkin data on a famous online social network platform called QQ. Inspired by [25], we combine the most visited check-in location during the night and collected home location to calculate the latitude and longitude of a person's home. Among 32,443 volunteers, 4,509 of them reported at least one socioeconomic attribute and agreed to share their home location for research purposes.…”