Powerful new AI can predict people's attitudes towards vaccines

This predictive system uses a small amount of data from demographics and personal judgments such as risk and loss aversion.

The findings provide the framework for a new technology that has wide applications in predicting mental health and could lead to more effective public health campaigns.

A team led by researchers at the University of Cincinnati and Northwestern University created a predictive model using an integrated mathematical system that describes regular patterns in machine learning reward and aversion judgments.

“We used a small number of variables and minimal computational resources to make our predictions,” said lead author Nicole Weick, a senior research fellow in the University of California's College of Engineering and Applied Sciences.

“COVID-19 is unlikely to be the last pandemic for decades to come. New forms of predictive AI in public health can help prepare hospitals to predict vaccination rates and subsequent infection rates.” It can be a valuable tool.”

This research Medical Internet Research Journal Public Health and Surveillance.

Researchers surveyed 3,476 adults across the United States in 2021 during the coronavirus disease (COVID-19) pandemic. At the time of the study, it had been more than a year since the first vaccines became available.

Respondents provided information such as where they live, income, highest level of education completed, ethnicity, and access to the Internet. Respondent demographics reflected U.S. demographics based on U.S. Census Bureau statistics.

Participants were asked whether they had received any of the available COVID-19 vaccines. About 73% of respondents said they had been vaccinated, slightly more than 70% of the population vaccinated in 2021.

They were also asked whether they routinely followed four recommendations to prevent the spread of the virus: wearing a mask, social distancing, washing hands, and not gathering in large groups.

Participants were asked to rate how much they liked or disliked a set of 48 randomly arranged pictures on a 7-point scale from 3 to -3. These photos are from the International Affective Picture Set, a large set of emotionally stimulating color photographs divided into his six categories: Sports, Disasters, Cute Animals, Aggressive Animals, Nature, and Food. It was something.

Weick said the goal of the exercise was to quantify the mathematical characteristics of people's judgments when observing mild emotional stimuli. Measures for this task include factors well known to behavioral economists and even gamblers, such as risk aversion (the point at which one is willing to accept a potential loss for a potential reward) and an aversion to loss. Contains concepts that are This refers to the willingness to avoid risks, such as by purchasing insurance.

“The framework for determining what is rewarding and what is aversive is the basis for how we make medical decisions,” said co-senior author and University of California says computer science professor Hans Breiter. “A seminal 2017 paper hypothesized the existence of a standard model of the mind. Using a small set of variables from mathematical psychology to predict medical behavior would support such a model. This collaborative team's research provides such support, arguing that the mind is a set of equations similar to those used in particle physics.

Judgment variables and demographics were compared between vaccinated and unvaccinated respondents. We used three machine learning approaches to test the extent to which respondents' judgment, demographics, and attitudes toward COVID-19 precautions predicted whether they would get vaccinated.

This study demonstrates that artificial intelligence can make accurate predictions about human attitudes with surprisingly little data or relying on expensive and time-consuming clinical assessments.

“We found that a small number of demographic variables and 15 decision variables were able to predict vaccine uptake with moderate to high accuracy,” the study said. “In the era of big data machine learning approaches, the current study provides an argument for using fewer, but more interpretable variables.”

“This study is anti-big data,” said co-senior author Ageros Katsagelos, an endowed professor of electrical engineering and computer science at Northwestern University. “It works very easily. It doesn't require advanced calculations, it's cheap, and anyone with a smartphone can apply it. We call it computational cognitive AI. Perhaps , we will also see other applications for changes to the system.'' A ruling will be rendered in the very near future. ”