Google has released its annual report outlining the results of its efforts to combat Search spam over the past year. It introduces its AI-based spam-prevention system called SpamBrain.
SpamBrain was launched in 2018, but this is the first time Google has written about it externally with its anti-spam measures.
Google says SpamBrain was built to be a “robust and evolving platform to address all types of abuse”, with the ability to identify disruptive and malicious behaviors among billions of web pages.
We caught 200 times more spam sites in 2021 compared to when we first started nearly two decades ago, thanks, in part, to our AI-based spam-prevention system called SpamBrain.
The search engine credits SpamBrain with identifying nearly six times more spam sites in 2021 than in 2020, resulting in:
- a 70% reduction in hacked spam
- a 75% redacted in gibberish spam on hosted platforms
- more than 99% of searches spam-free
Beyond traditional web spam
Google says it also made “significant progress” in 2021 in fighting link spam, scams and online harassment.
- A link spam update was released in July 2021, to tackle unnatural links more broadly across multiple languages and prevent them from affecting search quality.
- Several algorithm updates were rolled out as part of efforts to reduce scams, resulting in a 40% reduction in scammy results.
- SpamBrain was extended to address online harassment, reducing the prominence of sites with exploitative removal practices for name queries, such as charging for content removal.
Reducing ranking manipulation
Google also took measures to tackle behaviors that only just avoid violating its quality guidelines, but are still “manipulative in nature and degrade the user experience”.
It gave the example of its two product reviews updates in April and December 2021, which significantly reduced low grade reviews and increased prominence for high quality ones.