nlptoolkit-spellchecker

Turkish Spell Checker
============

This tool is a spelling checker for Modern Turkish. It detects spelling errors and corrects them appropriately, through its list of misspellings and matching to the Turkish dictionary.

Video Lectures
============

For Developers
============

You can also see Java, Python,
Cython, Swift,
C++, or C# repository.

Requirements

* Node.js 14 or higher
* Git

$3

To check if you have a compatible version of Node.js installed, use the following command:

node -v

You can find the latest version of Node.js here.

$3

Install the latest version of Git.

Npm Install

npm install nlptoolkit-spellchecker

Download Code

In order to work on code, create a fork from GitHub page.
Use Git for cloning the code to your local or below line for Ubuntu:

git clone

A directory called util will be created. Or you can use below link for exploring the code:

git clone https://github.com/starlangsoftware/turkishspellchecker-js.git

Open project with Webstorm IDE

Steps for opening the cloned project:

* Start IDE
* Select File | Open from main menu
* Choose SepllChecker-Js file
* Select open as project option
* Couple of seconds, dependencies will be downloaded.

Detailed Description
============

+ Creating SpellChecker
+ Spell Correction

Creating SpellChecker

SpellChecker finds spelling errors and corrects them in Turkish. There are two types of spell checker available:

* SimpleSpellChecker

* To instantiate this, a FsmMorphologicalAnalyzer is needed.

let fsm = FsmMorphologicalAnalyzer();
let spellChecker = SimpleSpellChecker(fsm);

* NGramSpellChecker,

* To create an instance of this, both a FsmMorphologicalAnalyzer and a NGram is required.

* FsmMorphologicalAnalyzer can be instantiated as follows:

let fsm = FsmMorphologicalAnalyzer();

* NGram can be either trained from scratch or loaded from an existing model.

* Training from scratch:

let corpus = Corpus("corpus.txt");
let ngram = NGram(corpus.getAllWords(), 1);
ngram.calculateNGramProbabilities(new LaplaceSmoothing());

There are many smoothing methods available. For other smoothing methods, check here.
* Loading from an existing model:

let ngram = NGram("ngram.txt");

For further details, please check here.

* Afterwards, NGramSpellChecker can be created as below:

let spellChecker = NGramSpellChecker(fsm, ngram);

Spell Correction

Spell correction can be done as follows:

let sentence = new Sentence("Dıktor olaç yazdı");
let corrected = spellChecker.spellCheck(sentence);
Console.log(corrected);

Output:

Doktor ilaç yazdı