This module compares two strings and returns a value on how similar they are from 0 - 1 (using a normalized levenshtein distance). The most basic use is:

> const {similarity} = require('how-similar')
> similarity('Dublin', 'Doblin')
0.833

The module can also compare a string to an array of strings to find which one is the most similar one:

> similarity('Dublin', ['Doblin', 'Devlin', 'Duuublin', 'Dublin'])
[ { key: 'Dublin', value: 1 },
  { key: 'Doblin', value: 0.875 },
  { key: 'Devlin', value: 0.75 },
  { key: 'Duuublin', value: 0.75 } ]

You can install with [npm]:

$ npm install --save how-similar

The module provides functions to deal with string / string comparison and string / array comparison. Still the easiest way to use it is to import similarity:

> const {similarity} = require('how-similar')

Then as shown above you can pass two strings or an string and an array of strings. You can also pass a third parameter with options (it will be explained next).

The following use cases can be solved with the module:

1. How similar are two different strings:

> similarity('Dublin', 'Doblin')
0.833

2. What is the most similar string in an array to another string:

> similarity('Dublin', ['Doblin', 'Devlin', 'Duuublin', 'Dublin'])
[ { key: 'Dublin', value: 1 },
  { key: 'Doblin', value: 0.875 },
...

3. How similar ignoring the case ({ignorecase: true}):

> similarity('Dublin', 'dUblIn', {ignorecase: true})
1

4. How similar stemming the words ({stem: true}):

> similarity('This is Dublin', 'Thi is Dublin', {stem: true})
1

5. How similar with exceptions ({except: [word1, word2, ...]})

> similarity('This is Dublin in Ireland', 'This is Dublin', {except: ['in', 'Ireland']})
1

Copyright © 2019, Juan Convers. Released under the MIT License.