Live
- Atkinson, Mendis, Ayub, Joseph named ICC Men’s Emerging Cricketer of the Year nominees
- Wanaparthy collector directs doctors to be available all time
- Parbhani custodial death: Prakash Ambedkar demands solatium, survey of victims of police brutality
- Indiramma illu survey should be completed in time, collector
- Excise police seize 10 litres of adulterated liquor
- AAP workers protest outside BJP leader Parvesh Verma's residence; latter distributes snacks
- I'll move the High Court, want written reply why I'm not getting Khel Ratna Award: Yogesh Kathuniya
- Malladi Venkata Krishnamurthy Lauds 'Master of Suspense Hitchcock'
- ACB hands over key documents to ED in the Formula E car race case
- How to Use Meta AI on WhatsApp for Creative Interactions
Google explains Maps review moderation as Yelp reports several violations
Google explains Maps review moderation as Yelp reports several violations
Google has explained how it keeps user-created reviews on Google Maps free from abuse and fraud in accompanying video and a new blog post. With machine learning as the first line of defense.
Google has explained how it keeps user-created reviews on Google Maps free from abuse and fraud in accompanying video and a new blog post. Like many platforms that deal with moderation at scale, Google says it uses a mix of automated machine learning systems and human operators.
The details come amid growing scrutiny of user reviews on sites like Google Maps and Yelp, where businesses have received poor reviews for implementing COVID-related health and safety measures (including mask and vaccination requirements), which are often beyond their control. Other reviews have criticized businesses for allegedly leading them to contract COVID-19 or failing to adhere to regular business hours during a global pandemic.
Earlier today, Yelp reported that it removed more than 15,500 reviews between April and December 2021 for violating its COVID-19 content guidelines, a 161% increase over the same period in 2020. In all, Yelp says it removed 70,200+ reviews on nearly 1,300 pages in 2021, many resulting from "review bombing" incidents where coordinated reviews are submitted from users who have not sponsored a business.
Google has shared that every review posted on Google Maps is reviewed by its machine learning system, which was trained on the company's content policies to weed out abusive or misleading reviews. This system can verify both the content of individual reviews and look for broader patterns, such as sudden spikes in one or five-star reviews, both from the account itself and from other reviews about the business.
Google says that human moderation helps for content that end-users and the companies themselves have flagged. Offending reviews may be removed and, in more serious cases, user accounts may be suspended and litigation initiated. "We've found that we need both the nuanced understanding that humans offer and the scale that machines provide to help us moderate contributed content, writes Google product lead for user-generated content Ian Leader.
It's an exciting look at the steps Google takes to keep Maps reviews usable. You can read more in the full blog post.
