Recently, Google published a blog post about how it keeps the information in the Maps app reliable. This application, which many people use as a tour guide in addition to its navigation feature, allows us to find sometimes the best restaurants and sometimes the best entertainment centers by sharing information – 20 million daily feedback – entirely from users.
Google Maps destroys fake information thanks to artificial intelligence!
Google includes artificial intelligence in many of its apps and services, including Maps. But in the aforementioned situation, it seems that there is no other solution. Because many comments, feedback and photos are shared every day, from the condition of the roads, the opening-closing hours of the restaurants, the quality of the order, the food sold in the hardware store or the Pet Shop.
Most of these shares are what we see. In other words, concrete data or opinions conveyed by real users. However, there are fake accounts, tradesmen who want to trip their competitors, or people who just want to make people victims by giving false information.
Google states that it receives 20 million feedbacks every day from this app alone. Therefore, this data needs to be filtered by a control mechanism. However, even just one day’s worth of data can take months to review when only operators try to do so.
This is exactly where we see the contributions of artificial intelligence and machine learning. Of course, there is operator support again. But the whole review happens automatically. On the one hand, it protects tens of thousands of businesses from attacks, and on the other hand, the victimization of people is prevented. Moreover, it is said that close to 100 million fake and false posts have been blocked in this way so far.
In a shared blog post on the subject, Google said for Maps:
In an ever-changing world, it’s important that Google Maps gives you the freshest, most up-to-date information possible. One way to do this is with contributed content. As with any platform that accepts contributed content, we must remain vigilant in our efforts to combat abuse and ensure that this information is accurate.
Thanks to a combination of machine learning and human operators, we continue to reduce the amount of content. Fake or abusive content seen on Maps is actually less than one percent of all content displayed on Maps.
We’ve seen individuals and groups try to use fake reviews as a tactic to harm local businesses. He was often encouraged by the public’s attention to differences of opinion.
What do you think about this issue? Don’t forget to share your feedback with us on the SDN Forum or in the comments!
42227
