RankBrain / DeepRank / FastSearch

These systems measure the semantic fit between a search query and a page: whether the meaning matches, not just the keywords. They measure relevance, not content quality. RankBrain was Google's first AI system for this purpose. DeepRank is a deeper language understanding layer. FastSearch accelerates this estimate for faster responses.

Learned mathematical representations (embeddings) of search queries and documents, plus behavioral calibration: which search results did users actually rate as satisfying?

Three systems for language understanding. RankBrain = an ML system for (novel) queries via embeddings (numeric representations of meaning). DeepRank = deeper language understanding (BERT-related). RankEmbedBERT = the BERT-based embedding field. They measure fit, not quality. [B fields, O RankBrain/BERT]

Expertise, Authority and Trust are each indirect. These systems flank all E-E-A-T dimensions through better understanding of meaning, but do not measure them directly. Experience has no bearing: similarity of meaning does not touch first-hand material.

Keyword stuffing (inserting as many search terms as possible) no longer works, because RankBrain recognizes whether the meaning fits. A text that genuinely answers the information need behind a query performs better. RankBrain is also a bridge to AI citation: pages that are semantically well positioned have better chances of appearing in AI Overviews (AIO).