Azimzada.StringSimilarity 1.0.3

Azimzada.StringSimilarity

A powerful .NET library for calculating string similarities and performing fuzzy searches. This library implements popular string similarity algorithms like Levenshtein Distance and Jaro-Winkler, making it perfect for tasks such as spell checking, search suggestions, and data matching.

Features

• 🚀 Fast and efficient implementation
• 📊 Percentage-based similarity scores (0-100%)
• 🔍 Multiple similarity algorithms
• 🎯 Easy to use API
• ✨ Support for .NET 9.0+

Installation

Install via NuGet Package Manager:

Install-Package Azimzada.StringSimilarity

Or via .NET CLI:

dotnet add package Azimzada.StringSimilarity

Current version: 1.0.3

Usage

First, add the namespace to your code:

using Azimzada.StringSimilarity;

Levenshtein Distance

The Levenshtein distance measures the minimum number of single-character edits required to change one string into another. The similarity score is returned as a percentage (0-100%).

double similarity = StringSimilarity.Levenshtein("hello", "hallo");
// Returns: 80.0 (because only one character is different)

double similarity2 = StringSimilarity.Levenshtein("kitten", "sitting");
// Returns a percentage showing how similar the words are

Jaro-Winkler Similarity

The Jaro-Winkler similarity is particularly good for short strings such as person names. It returns a percentage score (0-100%).

double similarity = StringSimilarity.JaroWinkler("martha", "marhta");
// Returns: A high percentage as these are very similar

double similarity2 = StringSimilarity.JaroWinkler("dwayne", "duane");
// Returns: A good match score due to common characters

Examples

Here are some real-world examples:

// Check if user input matches a correct word
double score = StringSimilarity.Levenshtein("programming", "programing");
if (score > 90)
{
    Console.WriteLine("Did you mean 'programming'?");
}

// Compare two names
double nameMatch = StringSimilarity.JaroWinkler("John Smith", "Jon Smith");
if (nameMatch > 85)
{
    Console.WriteLine("Names are likely to refer to the same person");
}

Algorithm Details

Levenshtein Distance

• Measures the minimum number of single-character edits needed to change one string into another
• Handles insertions, deletions, and substitutions
• Returns a normalized percentage score (0-100%)

Jaro-Winkler

• Designed for short strings like names and passwords
• Accounts for character transpositions
• Gives more favorable ratings to strings that match from the beginning
• Returns a percentage score (0-100%)

License

This project is licensed under the MIT License - see the LICENSE file for details.

Contributing

Contributions are welcome! Feel free to submit issues and pull requests.