Billboards advertising unhealthy food are concentrated in poorer听areas and areas with a higher proportion of overweight children in Liverpool, UK. These findings may also apply across the country.
Using a combination of artificial intelligence and street-view images, Mark Green and his colleagues at the University of Liverpool mapped the content and geographical location of more than 10,000 outdoor adverts in the city.
This kind of data is often held by private firms and wasn鈥檛 previously available publicly, says Green.
Advertisement
He cycled on Liverpool鈥檚 streets between 14 and 18 January, wearing a 360-degree camera that was programmed to take images twice a second.
The camera collected more than 26,600 street-level images. To analyse them, the researchers first used an existing machine-learning algorithm that was trained to isolate advertisements from the surrounding environment.
They also trained another algorithm on hundreds of thousands of existing advertisements in four categories: unhealthy food, alcohol, gambling-related ads and others such as those for cars. They then tasked this algorithm with classifying the collected ads into the four categories.
It identified 1335 food ads, 217 alcohol ads and 149 gambling ads. The researchers then mapped the location of the ads and correlated their placement with maps of UK neighbourhood demographics that include measures such as income, employment, education and health.
They found that the unhealthy food ads were concentrated in the most deprived neighbourhoods, areas where people are more likely to lack financial and other social resources, such as access to health and education. These neighbourhoods tend to be located closer to large roads, and are known to have a higher percentage of overweight primary school-aged children.
The unequal exposure to junk food advertising may result from the least deprived areas in Liverpool being leafy suburbs with few billboards, says Green. This trend is likely to be repeated in other cities. The team also found a concentration of food, gambling and alcohol ads around university student areas.
Previous research suggests that junk food advertising on television and social media can nudge children towards unhealthier food choices, but the role of outdoor advertising is less clear.
The research may help city councils and institutions to regulate unhealthy advertising. Since February last year, for example, junk food ads have been banned on the Transport for London network, as a measure to combat childhood obesity.
The advantage of using AI and street-view images is cost efficiency, says Green. 鈥淵ou can get a handle on the exposure in different contexts and different cities,鈥� he says.
Reference:
