Source: Harish Jonnalagadda / Android Central
While we’re still in the midst of the global COVID-19 pandemic, people around the world have begun traveling again as we get used to this new normal. Google Maps on iOS or any of the best Android phones is one of the primary apps people use to get around, and on Wednesday, Google provided some insight as to how it keeps reviews for businesses relevant, accurate, and authentic.
Machine learning is a big part of much that Google does, and that extends to how it handles reviews. As Google’s “first line of defense,” machine learning is able to pick up patterns to help determine if the content is legitimate. It does this by identifying offensive or off-topic content, inspecting accounts for any history of suspicious behavior, and checking for any activity that’s out of the ordinary.
However, Google product lead Ian Leader notes how human operators are necessary to help train the models to better understand the context of a review.
Training a machine on the difference between acceptable and policy-violating content is a delicate balance. For example, sometimes the word “gay” is used as a derogatory term, and that’s not something we tolerate in Google reviews. But if we teach our machine learning models that it’s only used in hate speech, we might erroneously remove reviews that promote a gay business owner or an LGBTQ+ safe space.
This machine learning training is the difference between a legitimate review going up in seconds or a policy-violating review being immediately blocked.
Google also encourages users to flag anything they find that may have slipped through the cracks and provides links for businesses and consumers to report policy-violating reviews. For instance, Google prohibits reviews that criticize companies for complying with health mandates related to COVID-19. These reports are reviewed by human operators who will respond accordingly.
To get more information on how Google Maps handles reviews, you can check out the full report.
