Google published a report this week on its efforts with reviews of places and businesses on Google Maps and the search service. According to big tech, 170 million fake reviews have been deleted. This total of removed comments was aided by a new algorithm from the company.
According to Google, this removal of 170 million fake reviews represents a 45% increase when compared to 2022. The machine learning-based algorithm can, among other things, identify the peak of 1 in 5 reviews at establishments even faster. These huge amounts of highly negative or positive reviews is one of the signs of fake reviews.
The main people responsible for these fake reviews are people hired by scammers. They offer a payment for each review made in certain locations you’ve probably received such a proposal on WhatsApp. The algorithm identifies patterns, whether in writing, accounts, or type of establishment, and refines the search to remove reviews linked to the scam.
According to Google, the reports made by the merchants themselves who are the targets of these reviews (including the fake 5-star ones) and the contact of those who received the proposal from the scammers help in the fight against fake reviews.
Google’s algorithm also works to identify reviews linked to high-profile cases. For example, if a restaurant goes viral with a negative experience and starts receiving bad reviews as a protest, the big tech system is able to identify the case. The moderation of these fake reviews is also supported by humans.
Consumption is one of the engines of Google services and browsers. Identifying fake reviews is important to provide users with a better shopping and leisure experience in commercial establishments. Last year, Mozilla released a tool for Firefox that identifies fake product reviews. The feature works similarly to Google’s algorithm, being based on machine learning and capable of identifying patterns in the texts of the authors of the reviews.
