The world is thriving on Artificial Intelligence (AI), and it has indeed become an essential component of our everyday life. The humanitarian sector is in the early days of harnessing this technology to expedite aid and humanitarian actions in much-needed areas.

In 2014, the International Organization for Migration (IOM) initiated the Missing Migrants Project (MMP) to document deaths and disappearances of people in the process of migration towards an international destination. The existing data on incidents involving migrant deaths are vastly incomplete. Among other sources, the project relies largely on media reports from across the world. The manual media monitoring process was a very resource-intensive task to address the issue to timely document migrant deaths and disappearances and the risks of migration, with the ultimate objective to save lives and establish coordinated international efforts on missing migrants. Since the initiation of the MMP, over 45,500 deaths and disappearances have been recorded, although the real number of lives lost during migration is presumed to be much higher.

The MMP partnered with Data Friendly Space to implement Machine Learning techniques to create a tool that would expedite the media monitoring process. In our track, we will discuss the background of the MMP project, some of the critical data collection challenges that the project faces and how AI is improving the documentation of migrant deaths and disappearances and contributing to bringing forward the issue to global stakeholders.

The speakers will also highlight the technical challenges they faced while training the ML models to extract the relevant information on missing migrants from a vast data set.