Abstract:Background: Youth experiencing homelessness face substance use problems disproportionately compared to other youth. A study found that 69% of youth experiencing homelessness meet the criteria for dependence on at least 1 substance, compared to 1.8% for all US adolescents. In addition, they experience major structural and social inequalities, which further undermine their ability to receive the care they need. Objective: The goal of this study was to develop a machine learning-based framework that uses the social media content (posts and interactions) of youth experiencing homelessness to predict their substance use behaviors (ie, the probability of using marijuana). With this framework, social workers and care providers can identify and reach out to youth experiencing homelessness who are at a higher risk of substance use. Methods: We recruited 133 young people experiencing homelessness at a nonprofit organization located in a city in the western United States. After obtaining their consent, we collected the participants' social media conversations for the past year before they were recruited, and we asked the participants to complete a survey on their demographic information, health conditions, sexual behaviors, and substance use behaviors. Building on the social sharing of emotions theory and social support theory, we identified important features that can potentially predict substance use. Then, we used natural language processing techniques to extract such features from social media conversations and reactions and built a series of machine learning models to predict participants' marijuana use. Results: We evaluated our models based on their predictive performance as well as their conformity with measures of fairness. Without predictive features from survey information, which may introduce sex and racial biases, our machine learning models can reach an area under the curve of 0.72 and an accuracy of 0.81 using only social media data when predicting marijuana use. We also evaluated the false-positive rate for each sex and age segment. Conclusions: We showed that textual interactions among youth experiencing homelessness and their friends on social media can serve as a powerful resource to predict their substance use. The framework we developed allows care providers to allocate resources efficiently to youth experiencing homelessness in the greatest need while costing minimal overhead. It can be extended to analyze and predict other health-related behaviors and conditions observed in this vulnerable community.

Can Machine Learning Create an Advocate for Foster Youth?

Predicting Successful Placements for Youth in Child Welfare with Machine Learning

Identifying Marijuana Use Behaviors Among Youth Experiencing Homelessness Using a Machine Learning-Based Framework: Development and Evaluation Study

A Conceptual Framework for Using Machine Learning to Support Child Welfare Decisions

A Data Driven Methodology for Social Science Research with Left-Behind Children As a Case Study.

A Data-Driven Simulation of the New York State Foster Care System

States of opportunity for youth aging out of foster care

Predicting suicidal thoughts and behavior among adolescents using the risk and protective factor framework: A large-scale machine learning approach

Cultivating a Supportive Sphere: Designing Technology to Increase Social Support for Foster-Involved Youth

Supporting Adolescent Engagement with Artificial Intelligence–Driven Digital Health Behavior Change Interventions

An Intelligent and Data-Driven Mobile Platform for Youth Volunteer Management using Machine Learning and Predictive Analytics

Case Study: Predictive Fairness to Reduce Misdemeanor Recidivism Through Social Service Interventions

Implementing and Evaluating Manualized Interventions in U.S. Organizations Serving Youth with Foster Care Experience: Lessons Learned from the Field

Youth as Peer Auditors: Engaging Teenagers with Algorithm Auditing of Machine Learning Applications

Predicting Adolescent Mental Health Outcomes Across Cultures: A Machine Learning Approach

The Potential of Diverse Youth as Stakeholders in Identifying and Mitigating Algorithmic Bias for a Future of Fairer AI

Predicting Homelessness among Emerging Adults Aging Out of Foster Care

Machine Learning-Based Research on the Adaptability of Adolescents to Online Education

Transforming the Judicial System: The Impact of Machine Learning on Legal Processes and Outcomes

A computational social science approach to understanding predictors of Chafee service receipt

Adolescent, Parent, and Provider Perceptions of a Predictive Algorithm to Identify Adolescent Suicide Risk in Primary Care