mirrors
/
StringDistances.jl
mirror of https://github.com/matthieugomez/StringDistances.jl.git
1
0
Fork 0
Code Issues Releases Wiki Activity

return

pull/7/head
matthieugomez 2018-05-17 11:27:31 -04:00
parent fb619ed4c5
commit 68c82cd40b
1 changed files with 5 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
README.md
View File

@ -93,16 +93,16 @@ The package includes distance "modifiers", that can be applied to any distance.
compare(TokenMax(RatcliffObershelp()),"mariners vs angels", "los angeles angels at seattle mariners")
#> 0.855
```
## Evaluate vs Compare
The function `evaluate` returns the litteral distance between two strings, with a value of 0 being completely similar.
## Compare vs Evaluate
The function `compare` returns a similarity score: a value of 0 means completely different and a value of 1 means completely similar.
In contrast, the function `compare` returns a similarity score: a value of 0 means completely different and a value of 1 means completely similar.
In contrast, the function `evaluate` returns the litteral distance between two strings, with a value of 0 being completely similar.
```julia
evaluate(Levenshtein(), "New York", "New York")
#> 0
compare(Levenshtein(), "New York", "New York")
#> 1.0
evaluate(Levenshtein(), "New York", "New York")
#> 0
```
## Which distance should I use?