RankBrain is a machine learning system, used by Google to guess the users’ intent behind searches and provide more relevant search results.
In a 2015 interview, Google commented that this artificial intelligence was the third most important signal in the ranking algorithm along with links and content. If RankBrain spots a word or phrase that it isn’t familiar with, the machine can make a guess as to what words or phrases might have a similar meaning and filter the result accordingly. In this way, Google handles effectively never-before-seen search queries or keywords, that – in 2015 – were 15% of all searches.
Studies showed how RankBrain better interpreted the relationships between words. This can include the use of stop words in a search query (such as articles, prepositions, conjunctions, etc), that were historically ignored previously by Google but are sometimes are highly important to fully understand the meaning or intent behind a person’s search query. It’s also able to parse patterns between searches that are seemingly unconnected, to understand how those searches are similar to each other.