News for hackers. Stuff that matters.
Re:Fuzzy string matching using cosine similarity by thomasahle on Monday September 21, 2015 @09:22AM (#10252030)
Many people comment that there are nicer string-distance measures.

However the real advantage of cosine distance is that you can perform dimensionality reduction (see e.g. page 38 of [1]). This allows you to work with very large documents efficiently and fuzzy. It also allows you to create efficient data structures for finding similar strings and much more.

[1]: http://web.stanford.edu/class/cs246/slides/04-lsh_theory.pdf

There are no comments on this comment yet.