Earlier this year I looked at a fascinating Yale project that used artificial intelligence to help tackle HIV. The work, which was documented in a recently published paper, aimed to improve our ability to detect ‘hotspots’ of HIV infection that can help public health teams identify areas of potential infection among people that don’t know they carry the virus.
It’s not the only project using technology to improve matters however, with a recent USC study seeing the creation of algorithms to help the spread of public health information to people with the HIV virus.
The method, which the team believe to be over 150% more effective than existing methods, works by identifying the most influential peer leaders to help spread the message to those in need.
Targeted messaging
The team created a couple of algorithms to try and predict which peer leaders would be most effective at spreading the various public health messages among a group of homeless youths. The algorithms, known as HEALER (which stands for Hierarchical Ensembling-based Agent which pLans for Effective Reduction in HIV spread) and DOSIM (which stands for Double Oracle for Social Influence Maximization), were tested over a seven month period.
Both proved effective in ultimately changing the behavior in the target groups, but there were subtle differences in performance. For instance, HEALER assumed that the young people would rank their friends in terms of closeness, but it’s not particularly easy to get this information. By contrast, DOSIM works without this information.
When analyzing existing methods of disseminating information, they found it to reach just 27% of the population. Both HEALER and DOSIM however managed to reach around 70% of people. What’s more, both approaches managed to convince more people to get tested, with HEALER proving especially effective.
“This paper shows the power of interdisciplinary research. AI algorithms for influence maximization were tested in the field to show the great benefits that accrue for low resource communities–research that would not get done without AI and Social Work researchers coming together to conduct it,” the authors say.
The next step is to further test the algorithms by expanding the cohort of participants they look to disseminate health information through.
