Unemployment is an important indicator of the performance of an economy.
One of the most promising rich new data sources is mobile phone data, which have the potential to deliver near realtime information of human behavior on individual and societal scale. A new Big Data Study provides the first confirmation that individual employment status can be predicted from standard mobile phone logs. The study addresses how machine learning models can be used to predict 18 categories of individual professions in a South-Asian developing country, and further predicts individual unemployment status with 70.4 percent accuracy. It shows how unemployment can be aggregated from individual level and mapped geographically at cell tower resolution, providing a promising approach to map labor market economic indicators. An important policy application of this work is the prediction of regional and individual employment rates in developing countries where official statistics is limited or non-existing, and support data collection on vulnerable populations which are frequently under-represented in official surveys.
Link to article: https://arxiv.org/pdf/1612.03870v1.pdf
In the news: How Metadata can Reveal what your job is
The above figure shows the geographical distributions of employment status and profession categories per base station in one of the larger Asian cities with over 1,500 cell towers and 18 million people. Employment rates are calculated by using the out-of-sample test set, aggregated and averaged to their respective home tower. Individual prediction accuracy and top three mobile phone predictors are given for unemployed, teachers/students, landlords, retired and clerks.