1

I'm searching for something that lets me fuzzy match in Python 3.6 without using the following libraries/packages that have been discarted (not my project, so I cannot make a decision over it unless I find a solution to the problems these libraries have):

  • tre library by laurikari (too many headaches during instalation and it has little to no documentation to help in that issue)
  • re/regex options for fuzzy matching (takes too long to process)
  • fuzzywuzzy (not tested officialy, making it not reliable for my project)

I've been searching for a solution but I seem to not find it so.... help?

thisismylasthope
  • 11
  • 1
  • fuzzywuzzy is likely the most widely-used python package for fuzzy matching. I doubt you're going to find anything that has been more extensively tested by the community. – dspencer Mar 12 '20 at 14:50

2 Answers2

1

Fuzzy matching relies on a distance metric between texts. You could just use a function that calculates the "levenshtein" (or any other metric) in your code (e.g. https://www.datacamp.com/community/tutorials/fuzzy-string-python). That way you don't need any external dependencies. Just loop the possible matches, calculate the distance and pick the one with the lowest distance (largest similarity).

If you would like an external library, just not one out of your "blacklist", try https://pypi.org/project/textdistance/

Gijs Wobben
  • 1,974
  • 1
  • 10
  • 13
0

You can use MatchKraft Python CLient. However,it only works for fuzzy matching companies. Here is the python library: https://github.com/MatchKraft/matchkraft-python. It is very easy to use.

user3643160
  • 1
  • 1