RecoLang

This project is based on a school project for the University of Nantes.

It is a java project wich recognizes different languages. With models for 10 languages, the program build files of bigrams and unigrams with frequencies of apparition. With those files, the program will analyse a sentence and find the language by using probabilities.

The purpose of the project was to run performance tests on files containing 22 000 sentences to recognize.

Construction

This project is build in two parts :

• the RecoLang part is used for the tests of performances. It's implementing the main algorithm of the program.
• the LanguageModel part is used for building the modelling of models for the languages. This part will evaluate probabilities and build files of n-grams for each language.

Languages

The languages used in the program are :

• Spanish
• Deutsch
• Hungarian
• Italian
• Latvian
• Dutch
• Polish
• Portuguese
• Swedish
• Estonian

The program can be used for building any language model if you used it with a corpus containing sentences of the language.

A part for recognize the unknown languages is also implemented.