Google wanted to share two new examples of how it is applying technology together with expert analysts to combat the publication of false contributions on Google Maps, even before they are published, given how useful this tool is throughout the world, used by more than a billion people a month.
The company is aware that the dynamics of fraudsters evolve over time, becoming increasingly complex, so its measures must also evolve according to these dynamics.
Machine learning models as important allies
Google points to the use of machine learning to be faster at detecting patterns of abuse, whose machine learning model update last year helped detect new abuse trends, including the sudden increase in abuse. commercial profiles whose web links ended in .design or .top, something difficult to detect manually in millions of profiles, but the analyst team confirmed that these websites are fake, leading to the suspension of said profiles.
And on the other hand, given the use of fraudulent phone numbers superimposed on images, Google launched a new machine learning model capable of recognizing numbers superimposed on images through the analysis of specific visual details and photo layouts, managing to detect and block most of these images before they get published.
The achievements made in 2022 against the increase in fraud
In addition to these two examples, Google has also wanted to share its progress in combating false contributions over the past year by combining machine learning models with advanced manual and automated techniques to deal with false content.
They point out that thanks to their updated models, they removed 20% more fake reviews than in 2021, most of them before they were even seen by the public. Using their techniques, they blocked or removed more than 200 million photos and 7 million videos that were blurry, low-quality, or otherwise violated their content policies.
They also managed to stop 20 million attempts to create fake Business Profiles, 8 million more than in 2021, and even applied protections to more than 185,000 businesses after suspicious activity and abuse attempts were detected.
To all this, Google also goes to court to take legal action against offenders, giving as an example the filing of a lawsuit that made it possible to successfully eliminate a group of scammers who came to impersonate Google through fraudulent commercial calls to try to get fake reviews.
Everything they have learned they also share with the FTC and government agencies around the world to deal with fake reviews and other misleading content. With what was published today by Google, we can now get an idea of ​​the work behind it so that we only access reliable information about the places we are interested in without falling into fraud perpetrated by malicious actors.
More information/image credit: Google
