Skip to content
No description, website, or topics provided.
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
src/main
.gitattributes
.gitignore
README.md
pom.xml

README.md

SpellChecker

For Developers

Requirements

Java

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

java -version

If you don't have a compatible version, you can download either Oracle JDK or OpenJDK

Maven

To check if you have Maven installed, use the following command:

mvn --version

To install Maven, you can follow the instructions here.

Git

Install the latest version of Git.

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 <your-fork-git-link>

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

git clone https://github.com/olcaytaner/SpellChecker.git

Open project with IntelliJ IDEA

Steps for opening the cloned project:

  • Start IDE
  • Select File | Open from main menu
  • Choose SpellChecker/pom.xml file
  • Select open as project option
  • Couple of seconds, dependencies with Maven will be downloaded.

Compile

From IDE

After being done with the downloading and Maven indexing, select Build Project option from Build menu. After compilation process, user can run SpellChecker.

From Console

Use below line to generate jar file:

 mvn install

SpellChecker

Maven Usage

<groupId>NlpToolkit</groupId>
<artifactId>SpellChecker</artifactId>
<version>1.0.1</version>
<properties>
    <maven.compiler.source>1.8</maven.compiler.source>
    <maven.compiler.target>1.8</maven.compiler.target>
</properties>
<repositories>
    <repository>
        <id>NlpToolkit</id>
        <url>http://haydut.isikun.edu.tr:8081/artifactory/NlpToolkit</url>
    </repository>
</repositories>
<dependencies>
    <dependency>
        <groupId>NlpToolkit</groupId>
        <artifactId>Corpus</artifactId>
        <version>1.0.0</version>
    </dependency>
    <dependency>
        <groupId>NlpToolkit</groupId>
        <artifactId>Dictionary</artifactId>
        <version>1.0.2</version>
    </dependency>
    <dependency>
        <groupId>NlpToolkit</groupId>
        <artifactId>MorphologicalAnalysis</artifactId>
        <version>1.0.2</version>
    </dependency>
    <dependency>
        <groupId>NlpToolkit</groupId>
        <artifactId>NGram</artifactId>
        <version>1.0.0</version>
    </dependency>
</dependencies>

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.

        FsmMorphologicalAnalyzer fsm = new FsmMorphologicalAnalyzer();
        SpellChecker spellChecker = new SimpleSpellChecker(fsm);   
      
  • NGramSpellChecker,

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

    • FsmMorphologicalAnalyzer can be instantiated as follows:

        FsmMorphologicalAnalyzer fsm = new FsmMorphologicalAnalyzer();
      
    • NGram can be either trained from scratch or loaded from an existing model.

      • Training from scratch:

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

      There are many smoothing methods available. For other smoothing methods, check here.

      • Loading from an existing model:

          try {
              FileInputStream inFile = new FileInputStream("ngram.model");  
              ObjectInputStream inObject = new ObjectInputStream(inFile);
              NGram ngram = (NGram<Word>) inObject.readObject();
          }catch (IOException | ClassNotFoundException e) {
              e.printStackTrace();
          }
        

      For further details, please check here.

    • Afterwards, NGramSpellChecker can be created as below:

        SpellChecker spellChecker = new NGramSpellChecker(fsm, ngram);
      

Spell Correction

Spell correction can be done as follows:

Sentence sentence = new Sentence("Dıktor olaç yazdı");
Sentence corrected = spellChecker.spellCheck(sentence);
System.out.println(corrected);

Output:

Doktor ilaç yazdı
You can’t perform that action at this time.