Explainable AI for Predicting Chronic Homelessness

Homelessness is a pressing issue. In Canada, more than 235,000 individuals experience homelessness every year and on any given night, between 25,000 and 35,000 people may be affected. The general goal of this project is to develop machine learning (ML)-based tools to help with the process of tackling the issue of chronic homelessness. More specifically, the goal is to 1) develop an ethical and unbiased machine learning-based solution to predict the total number of nights an individual will stay in a shelter six months in the future, 2) provide explanations about decisions made by the resulting ML model, and 3) gain knowledge about some of the internal and external factors that may affect the likelihood of recurrence of homelessness.

Faculty Supervisor:

Majid Komeili;Olga Baysal

Student:

Partner:

City of Ottawa

Discipline:

Computer science

Sector:

Public administration

University:

Carleton University

Program: