Google has said it blocked or removed over 170 million fake reviews that violated its policies in 2023. According to a recent blog post, the tech giant stated that this was over 45% more than in 2022.
It also removed more than 12 million fake business profiles from its search engine, which they say is thanks to a new machine learning algorithm. “Last year, we launched a new machine learning algorithm that detects questionable review patterns even faster,” the company said, adding that it does so by “examining longer-term signals on a daily basis.”
The California-based firm claimed that over 20 million community reviews of products and services in Search and Maps are added every day, which requires constant content moderation “to fight bad actors.”
Hence, their new technology reportedly helps identify surges, such as fake reviews or ad clicks, allowing a team of analysts to inspect patterns and refine the algorithms. As a result, five million fake reviews were removed in just a few weeks.
What’s more, Google said that it removed 14 million videos that violated its policies, which was double than the previous year. It reported that business owners were protected against more than two million attempts by malicious entities attempting to claim business profiles that did not belong to them.
Google’s battle against search engine issues
In 2023, the Alphabet-owned company said it had filed a lawsuit against an individual who had been posting fake reviews on Google Maps and fraudulently attempting to manipulate services for small businesses.
However, according to a recent study entitled “Is Google Getting Worse?,” researchers found that there had been an influx of junk content on its search engines, and that engineers were struggling to limit the number of SEO spam that was constantly entering and leaving the results.
Consequently, the Big Tech organization has stated that improvements were coming to the results page, from grouping links to comparison sites from across the web to query shortcuts designed to help users find what they need faster.
Featured image: