Projects

Project Kin

A composite image of the various platforms developed in the project, including a phone application, a desktop platform, and an IVR feature.

by Priyanka Pillai (MDE ’24) and Julius Stein (MDE ’24)

When conflict arises from humanitarian crises, families are invariably separated. Sometimes, this occurs intentionally if the family decides it is safer to leave the child in someone else’s care or in another location. Other times, this happens accidentally during the disarray and confusion following the conflict, which can cause a child to suddenly find themselves alone or with a group of travelers from their home who are not their family. Still at other times, this occurs during travel from one’s home to their destination either because of groups splitting apart or because of accidental separations. 

Current approaches involve the use of pen and paper tracking, grids of posters and billboards of faces, and direct networking in person or over the phone without the help of any organization. These not only have low success rates, but they frequently expose the vulnerable party to unwanted attention. We would like to see what other alternatives could potentially offer higher likelihoods of reconnection without the risk that current approaches typically take on. Additionally, few systems give power to the displaced parties to reconnect themselves and instead will often place power in the hands of organizations, which we hope to avoid.

A set of images containing physical prototypes made by refugees including a model figurines and pipe cleaner houses, as well as photos of their fabrication and presentation.

Our approach, Project Kin, addresses the problem of family separation in refugee camps through an accessible online platform for refugees that allows them to find relatives themselves, returning ownership of the process to the displaced people. Families can register with the platform before separation, when arriving at the camp, or after they have been separated from their family. Project Kin begins by prompting refugees to input information about themselves through a combination of text and audio, and generates a set of follow-up questions based on their responses.

Advisors: Martha Thompson (MIT D-Lab), Krzysztof Gajos (SEAS), Shuya Gong (SEAS), Kathleen Brandenburg (GSD), Karen Reuther (GSD), Jon Jal Dak (YSAT, OXFAM)

A composite image of cell phone screens with pipe-cleaner sculptures in the background.
Several screens used in the phone application including the onboarding instructions, profile creation process, and Kinnektor minigame.
A set of diagrams explaining the different ways that users of Project Kin can interact with computers, phones, posters, and non-profit workers.
An overview of the various touchpoints for the project, including a phone application, a desktop platform, and physical assets including posters and implementation plans.
A classroom in Uganda with refugees of differing ages holding a web of string.
Refugees in Uganda share their experiences of family separation.
Diagrams showing how a software system can identify familial and community connections.
A set of heatmaps showing the predicted similarity of users, highlighting matches which correctly identified sibling pairs, and highlighting regional similarity among the study participants.
A diagram comparing existing methods of family connection to kin.
An analysis of existing methods used to solve family separation, scoring each on several metrics and communicating visually through circle sizes.
A diagram showing the "tech pipeline" for the development of the Kin product including translation and text-to-speech services.
A technical overview of the backend detailing the platforms and development languages used in each component of the program.