Full text search is a technique that finds entries in a collection of text based on the presence of specific keywords and phrases.
For instance, consider a movie review site that lets users search for movies. Movie titles would benefit from full text search, since users are
likely looking for exact keywords like Harry and Potter.
Similarity search matches documents based on semantic meaning. In the movie review example, movie descriptions may benefit from similarity search.
Users that query for the boy who lived may be looking for Harry Potter even though these phrases share no common keywords.This is achieved through a technique called vector search. A vector is a fixed array of numeric values that captures
the semantic meaning of a piece of text. Vectors are typically generated by embedding models.
Many modern applications use a combination of full text and similarity search. This process is called hybrid search.
Typical hybrid search techniques involve calculating separate full text and similarity scores for the result set and combining the scores into a hybrid score.
