Google says it blocked over 100 million ‘abusive edits’ to its Google Maps platform in 2021, by using both human and machine learning models.
In a blog post on how it kept Google Maps reliable amid a torrent of malevolent activity last year, the company explained how fights abuse on the platform and keeps the fraudulent edits seen by users at under one per cent, despite receiving 20 million user edits every day.
Google says fraudsters are constantly trying to update legitimate business profiles with abusive
edits, for example, but a combination of its AI and human teams have managed to shut down most of them.
Google says it identified 7 million fake business profiles, of which 630,000 were reported directly. It says 12 million attempts to create fake profiles were stopped at the source, as were 8 million fraudulent attempts to claim legit profiles.
“Thanks to continued improvements in our machine learning, our technologies and teams disabled more than 1 million user accounts due to policy-violating activity, such as online vandalism or fraud,” the company added in the blog post.
Google also discussed its efforts to eradicate review bombing and says that 95 million policy-violating reviews were blocked or removed, many of which were designed to hurt local businesses “often spurred by public attention on differences of opinion” including on matters like Covid. One million reviews were removed after being reported to Google directly.
“In 2021, to prevent violative content from appearing on Google Business Profiles, we put protections on over 100,000 businesses after detecting suspicious activity and abuse attempts,” the company says.
“Local knowledge that our global community contributes is a huge part of what makes Google Maps more than a navigation tool. We’ll continue to invest in keeping this information fresh and reliable so you can discover information about the world around you.”