Clusters of Social Isolation Trajectories

Poster Number: P137
Presentation Time: 05:00 PM - 06:30 PM

Abstract Keywords: Machine Learning, Data Mining, Behavioral Change
Primary Track: Applications

Research has addressed social isolation (SI) and its health outcomes; however, machine learning methods have not been applied to detect SI trajectories. The Health and Retirement Study was utilized to examine SI trajectories. Interpolation and extrapolation were performed to handle unequal length of trajectories. K-means algorithm was employed to cluster SI. Findings show consistent SI scores over time for the majority. In contrast, one cluster showed worsening scores, possibly due to a decline in social participation.

Speaker(s):
Ghaida Alsadah, Health Services Research, PhD
George Mason University

Author(s):
Ghaida Alsadah, Health Services Research, PhD - George Mason University; Janusz Wojtusiak, PhD - George Mason University;