Home » YouTube’s fight against spam: the news to block harassing comments

YouTube’s fight against spam: the news to block harassing comments

by admin
YouTube’s fight against spam: the news to block harassing comments

Over 1.2 billion comments removed in the last 3 months of 2021, creator protests, 30% of messages posted under the most successful videos that would be unwanted: undoubtedly YouTube has a problem with spam. And now the platform has decided to approach it more carefully.

In response to the growing number of complaints from major youtubers such as Linus Tech, Marques Brownlee and Seán McLoughlin, the company has confirmed to Italian Tech that “we are working to improve the quality of the comments on YouTube through new features that will help creators interact with their community and set the tone for conversations on their channels ”.

Climate change

Google’s idea: machine learning and AI to predict the effects of floods

by Emanuele Capone


A new filter under the heading Increase Severity

What does it mean? What are these “new features”? From what we understand in the part dedicated to comments on the YouTube Help page, the authors will be able to select the Increase Strictness (in Italian, Increase Severity) item in their YouTube Studio dashboard if they want “an even higher level of protection”, so as to “retain more potentially inappropriate comments among those waiting for revision”. In practice, activating this setting “a more stringent threshold will be used for the detection of potentially inappropriate content in comments “.

The feature is still labeled Experimental, and YouTube allegedly began testing it in December last year: the problem is that at the moment it is based on the intervention of the creators, who will have to manually choose whether to approve, block or report every single comment among those held for review. The benefit, as explained, is that “these review actions help our technology to improve in identifying future comments.” The idea, in short, is that the artificial intelligence that supports humans in moderation gradually learns to work better and better, gradually becoming more and more autonomous.

See also  Serie A, Atalanta-Verona 3-1: Gasperini remains attached to Europe
A Google Opinion Rewards question about comments in videos

A Google Opinion Rewards question about comments in videos

Thus the AI ​​learns to recognize spam

There is another way YouTube software can distinguish quality comments from garbage, and we can help too: those who use l’app Opinion Rewards (who gives away money to spend on the Play Store in exchange for answers on various topics) will have noticed that for a couple of weeks, an increasing number of questions have been dedicated to this topic. Specifically, after seeing any video on YouTube, you may be asked to rate the quality of some of the comments that accompany it. So, if Google fails to understand for yourself that “weggpofqsx” is a useless and bad comment (example above)we can explain it to you.

Obviously, all this is not enough and will hardly be enough. And YouTube folks know: “We will continue to adapt our systems to stay up to date on new spammers’ tactics, which are always constantly evolving, ”they told us. And yet, the important thing is to start.

You may also like

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.

This website uses cookies to improve your experience. We'll assume you're ok with this, but you can opt-out if you wish. Accept Read More

Privacy & Cookies Policy