Skip to Main Content
You are now in the main content area

STRATEGIC FORESIGHT: Analyzing migration trends using machine learning

View of immigration control at Changi International Airport in Singapore

Yousef Khalifa Aleghfeli (Project Manager), Vaidehi Atodaria (Researcher), Sarah Hoyos-Hoyos (Methodological Advisor)

This project aims to build a reliable methodology that can predict migration trends using real-time data from various open-access sources. This involves identifying drivers of mobility, cross-border migration and internal displacement to anticipate shifts in mobility patterns across regions and countries using real-time data analysis. The project aims to provide evidence-based insights that can improve responses to forced displacement and inform strategic planning around migration management.

  1. In which countries are mobility dynamics likely to change?
  2. What key drivers (biophysical; socioeconomic; sociopolitical) are likely to impact mobility trends?
  3. How will these key drivers impact inflow/outflow of migration, and by how much?

The project is developed in response to the challenges facing governments and international organizations in optimizing resource allocation and decision-making to address evolving global migration trends. Migration dynamics, driven by factors such as political instability, economic stress, social tensions, and environmental changes, require real-time monitoring and analysis. As such, there is a growing need for reliable and evidence-based insights to anticipate and respond effectively to these mobility shifts. To achieve this, the project leverages the real-time and open-source data for extracting and analyzing factors that impact migration trends. The project focuses on biophysical, socioeconomic, and sociopolitical events likely to influence mobility patterns such as cross-border migration and internal displacement. By developing reliable methodologies grounded in data science, the project aims to inform strategic foresight and planning around forced displacement and migration management.

The project uses a four-phase methodology:

  • Phase 1 scans and leverages open-source and real-time data to extract mobility-impacting events.
  • Phase 2 interprets these events (biophysical; socioeconomic; sociopolitical) using cross-impact analysis, machine learning, and natural language processing.
  • Phase 3 aggregates the events into index scores, and tracks changes in these scores over time to identify key mobility drivers in different countries and regions.
  • Phase 4 applies geo-spatial modelling to visualizing the index scores and map key mobility drivers.

The project is in Phase 1.

Dec. 2025

CERC Migration

Partners

IOM Global Data Analytics

cross-border migration; internal displacement; data science