Introduction

The smart-match module contains functions for calculating strings/sets similarity.

Concept

1. similarity: A value in a range of [0, 1], which represents how similar the two strings are. The larger the value, the more similar the two strings are.

2. dissimilarity: A value in a range of [0, 1], which represents how dissimilar the two strings are. The larger the value, the more dissimilar the two strings are. For a pair of strings, similarity = 1 - dissimilarity

3. distance: How far the two strings are. Notice that not all the methods support distance method.

4. score The larger the score, the more similar the two strings are. Notice not all the methods have score method.

We support three levels of string matching.

1. char: Similarity computation based on characters in the strings.

2. term: Similarity computation based on terms in the strings.

3. gram: Similarity computation based on q-grams in the strings.

Methods

We support the following methods.

Abbreviation	Full name	similarity	dissimilarity	distance	score
LE(Default)	Levenshtein	✅	✅	✅	❌
ED	EuclideanDistance	✅	✅	✅	❌
DL	Damerau Levenshtein	✅	✅	✅	❌
BD	Block Distance	✅	✅	✅	❌
cos	Cosine Similarity	✅	✅	❌	❌
TC	TanimotoCoefficient	✅	✅	❌	❌
dice	Dice Similarity	✅	✅	❌	❌
simon	SimonWhite	✅	✅	❌	❌
LCST	LongestCommonSubstring	✅	✅	✅	✅
LCSQ	LongestCommonSubSequence	✅	✅	✅	✅
OC	OverlapCoefficient	✅	✅	❌	❌
GOC	GeneralizedOverlapCoefficient	✅	✅	❌	❌
jac	Jaccard	✅	✅	❌	❌
gjac	GeneralizedJaccard	✅	✅	❌	❌
HD	HammingDistance	✅	✅	✅	❌
jaro	Jaro	✅	✅	❌	❌
JW	JaroWinkler	✅	✅	❌	❌
NW	NeedlemanWunch	✅	✅	❌	✅
SW	SmithWaterman	✅	✅	❌	✅
SWG	SmithWatermanGotoh	✅	✅	❌	✅
MK	MongeElkan	✅	✅	❌	❌

Installation

pip install smart-match

Usage

import smart_match
print(smart_match.similarity('hello', 'hero'))
print(smart_match.dissimilarity('hello', 'hero'))
print(smart_match.distance('hello', 'hero'))

Output:

0.6
0.4
2

Check Wiki for more details.

License

smart-match is a free software. See the file LICENSE for the full text.

Authors