I'm trying to join datasets with names that partially match, ex. "John Smith" vs. "Jonathan Q. Smith". I'm familiar with the fuzzy joins package and am wondering if there are other libraries out there that can do better. Thanks!
1 Answer
I've never used it, but perhaps the RecordLinkage package, a research project from 2010 described in an article in R-Journal. Authors recommend to keep an eye on the NaturalLanguageProcessing CRAN Task View.
By the way, if what you want to do is "Address deduplication" I hear that many products apparently exist for this in the "Customer relationship management" space, or in the Email Marketing space