The effects of the so-called "refugee crisis" of 2015-16 continue to dominate much of the European political agenda. Migration flows were sudden and unexpected, exposing significant shortcomings in the field of migration forecasting and leaving governments and NGOs unprepared. Migration is a complex system typified by episodic variation, underpinned by causal factors that are interacting, highly context dependent and short-lived. Correspondingly, migration nowcasts rely on scattered low-quality data and much-needed forecasts are local and inconsistent. Here we describe a data-driven adaptive system for forecasting asylum applications in the European Union (EU), built on machine learning algorithms that combine administrative data with non-traditional data sources at scale. We exploit three tiers of data: geolocated events and internet searches in countries of origin, detections at the EU external border, and asylum recognition rates in the EU, to effectively forecast individual asylum-migration flows up to four weeks ahead with high accuracy. Uniquely our approach a) models individual country-to-country migration flows; b) detects migration drivers early onset; c) anticipates lagged effects; d) estimates the effect of individual drivers; and e) describes how patterns of drivers shift over time. This is, to our knowledge, the first comprehensive system for forecasting asylum applications based on an unsupervised algorithm and data at scale. Importantly, this approach can be extended to forecast other migration social-economic indicators.