Predicting Chronic Homelessness: The Importance of Comparing Algorithms using Client Histories

  title={Predicting Chronic Homelessness: The Importance of Comparing Algorithms using Client Histories},
  author={Geoffrey G. Messier and Caleb John and Ayush Malik},
This paper investigates how to best compare algorithms for predicting chronic homelessness for the purpose of identifying good candidates for housing programs. Predictive methods can rapidly refer potentially chronic shelter users to housing but also sometimes incorrectly identify individuals who will not become chronic (false positives). We use shelter access histories to demonstrate that these false positives are often still good candidates for housing. Using this approach, we compare a… Expand

Tables from this paper


Interpretable Machine Learning Approaches to Prediction of Chronic Homelessness
The model, HIFIS-RNN-MLP, incorporates both static and dynamic features of a client's history to forecast chronic homelessness 6 months into the client's future, achieving state-of-the-art performance and improved stakeholder trust of what is usually a "black box" neural network model through interpretable AI. Expand
Predictive modeling of housing instability and homelessness in the Veterans Health Administration
Predictive models based on medical record data can identify Veterans likely to report housing instability and homelessness, making the HSCR screening process more efficient and informing new engagement strategies. Expand
Efficient targeting of homelessness prevention services for families.
A model to target homelessness prevention services to families more efficiently in New York City found no evidence that some families were too risky to be helped or that specific risk factors were particularly amenable to amelioration. Expand
Allocating Interventions Based on Predicted Outcomes: A Case Study on Homelessness Services
Using data from the homeless system, a counterfactual approach is used to show potential for substantial benefits in terms of reducing the number of families who experience repeat episodes of homelessness by choosing optimal allocations to a fixed number of beds in different types of homelessness service facilities. Expand
Prioritizing Homeless Assistance Using Predictive Algorithms: An Evidence-Based Approach
This predictive analytic model prioritizes high-cost individuals for whom the solution of housing costs less than the problem of homelessness. Cost offsets from reduced service use after high-costExpand
Targeting Services to Individuals Most Likely to Enter Shelter: Evaluating the Efficiency of Homelessness Prevention
Successful strategies for homelessness prevention must efficiently target people at greatest risk for experiencing homelessness. We developed a model to target homelessness prevention services toExpand
Evidence From the Past: AI Decision Aids to Improve Housing Systems for Homeless Youth
There is much potential for effective human-machine collaboration in the context of housing allocation and this work determines whether the current rubric for prioritizing youth for housing assistance can be used to predict youth’s homelessness status after receiving housing assistance. Expand
Applications of Machine Learning Methods to Predict Readmission and Length-of-Stay for Homeless Families: The Case of Win Shelters in New York City
ABSTRACT New York City faces the challenge of an ever-increasing homeless population with almost 60,000 people currently living in city shelters. In 2015, approximately 25% of families stayed longerExpand
Applying Cluster Analysis to Test a Typology of Homelessness by Pattern of Shelter Utilization: Results from the Analysis of Administrative Data
Results suggest that program planning would benefit from application of this typology of homelessness, possibly targeting the transitionally homeless with preventive and resettlement assistance, the episodically homeless with transitional housing and residential treatment, and the chronically homeless with supported housing and long-term care programs. Expand
Predictors of Adolescents' First Episode of Homelessness Following Substance Use Treatment.
Adolescents who were older, male, reported more victimization experiences, mental health problems, family problems, deviant peer relationships, and substance use problems were more likely to report experiencing homelessness. Expand