JMIR Publications

ABSTRACT

Background:

Online mental health communities (OMHCs) are an effective and accessible channel to give and receive social support for individuals with mental and emotional issues. However, a key challenge on these platforms is finding suitable partners to interact with given that mechanisms to match users are currently underdeveloped or highly naive.

Objective:

In this paper, we collaborate with one of the world's largest OMHCs to contribute the application of agent-based modeling for the design of online community matching algorithms. We develop an agent-based simulation framework and explore how it can uncover trade-offs in different matching algorithms.

Methods:

We use a dataset spanning January 2020 to April 2022 to create a simulation framework based on agent-based modeling that replicates the current matching mechanisms of our research site. After validating the accuracy of this simulated replication, we use this simulation framework as a “sandbox” to test different matching algorithms based on the deferred-acceptance algorithm. We compare and contrast trade-offs among these different matching algorithms based on various metrics of interest such as chat ratings and matching success rates.

Results:

Our study contributes the novel application of agent-based simulation to matching in online health communities, and our simulation findings suggest that various tensions and goals emerge through different algorithmic choices for these communities. For example, we found that higher chat ratings and lower blocking frequency occurs with matching people using just topic(s) of interest for discussion, compared to matching based on just demographics or first-come-first-serve methods. We also found some trade-offs in hard filter-based approaches to prioritize the protection of marginalized groups, and that other algorithms can actually improve the experience of both minority and majority groups.

Conclusions:

Agent-based modeling can reveal significant design considerations in the OMHC context, including trade-offs in various outcome metrics and the potential benefits of algorithmic matching on marginalized communities.