Development of a Machine Learning Model Using Multiple, Heterogeneous Data Sources to Estimate Weekly US Suicide Fatalities

Abstract

The proposed ensemble machine learning framework reduces the error for annual suicide rate estimation to less than one-tenth of that of current forecasting approaches that use only historical information on suicide deaths. These findings establish a novel approach for tracking suicide fatalities in near real time and provide the potential for an effective public health response such as supporting budgetary decisions or deploying interventions.