Google is purchasing a company that provides city guides using image recognition and neural technology. Called Jetpac, it uses the data to recommend places that are the most popular, have the best views or are considered the happiest.
“We can spot lipstick, blue sky views, hipster moustaches and more, through advanced image processing on billions of photos,” reads Jetpac's homepage.
The program works by analyzing public images with location data shared through Flickr, Instagram, and other photo sharing websites. It then extracts key information about them.
Neural networks are essentially a group of algorithms which basically mimic the functioning of brain systems. They can be trained to recognize specific elements in images or text. They can also flag elements that are reoccurring or absent.
To train a neural network, programmers would show it a large number of photos and tell the algorithms when they are correct or incorrect. The network adjusts itself along the way until it is correct in the majority of instances.
Jetpac has previously identified Belfast as UK's happiest city, which they based on public Instagram photos.
Learn more about Jetpac in the video below.