Another snippet from Marie Haynes interview of John Mueller was around BERT. The short snippet, as Glenn Gabe nicely summed up, is that BERT helps Google understand content better but it does not help Google with evaluating quality.
These two tweets from Glenn sum it up nicely:
More from John: "Where BERT does overlap with quality is when people write in a way where it's impossible for BERT to understand what you're trying to get across. It's not making quality decisions. It just can't understand what the content is about." https://t.co/8Za6559FR1
— Glenn Gabe (@glenngabe) June 4, 2020
This is all around the 27 minute mark into the podcast interview.
John said "the one thing where people sometimes get thrown off is with regards to BERT, the BERT algorithms, all of those things. Those are essentially algorithms to understand content better. It’s not so much to understand the quality of the content but more to understand what is this content about, what is this sentence about, what is this query about to figure out what are the different entities that might be involved here and how are they being tied in here."
This clarification is nice and hopefully will help our industry understand the difference between BERT and normal quality algorithms.
Forum discussion at Twitter.
