Big Data for Social Sciences: Measuring patterns of human behavior through large-scale mobile phone data

The present study provides the first evidence that illiteracy can be reliablypredicted from standard mobile phone logs. By deriving a broad set of mobile phoneindicators reflecting users’ financial, social and mobility patterns we show howsupervised machine learning can be used to predict individual illiteracy in an Asiandeveloping country, externally validated against a large-scale survey. On averagethe model performs 10 times better than random guessing with a 70% accuracy.Further we show how individual illiteracy can be aggregated and mappedgeographically at cell tower resolution. Geographical mapping of illiteracy is crucialto know where the illiterate people are, and where to put in resources. Inunderdeveloped countries such mappings are often based on out-dated householdsurveys with low spatial and temporal resolution. One in five people worldwidestruggle with illiteracy, and it is estimated that illiteracy costs the global economymore than $1 trillion dollars each year [1]. These results potentially enable cost-effective, questionnaire-free investigation of illiteracy-related questions on anunprecedented scale.