BERT
A transformer language model Google applied to search from 2019, improving understanding of word order, prepositions and context within…
Algorithms
RankBrain
Google's first machine-learning ranking system, introduced in 2015, which helps interpret how queries relate to concepts, especially unfamiliar ones.
In full
RankBrain converts queries into vector representations so that never-before-seen searches can be matched to related known concepts. Google describes it as one of several AI systems used in ranking, working alongside neural matching, BERT and MUM rather than replacing them. There is nothing to optimise for directly; its practical implication is that content covering concepts thoroughly outperforms content matching phrases exactly.
Example
An unusually phrased query returns results that share no wording with it but address the same underlying need.
Related terms
Neural matching
A Google system that uses neural networks to understand how queries relate to page concepts, bridging vocabulary gaps between what users…
Semantic search
Retrieval based on the meaning of a query and documents rather than literal keyword matching, using entities, context and learned…
Embedding
A numerical vector representing the meaning of text, an image or another item, positioned so that semantically similar items lie close…
Google ranking systems
Google's documented list of the systems that generate rankings, distinguished from one-off 'updates' to those systems, and including…