language deteciton program written in java
naive bayes classifier.
features are character level ngrams in the text.
trained model using bag of n-grams
first parameter is phase "train/test", second parameter is model path, third parameter is inputData set, if it is test, you can set the fourth parameter to be the order of ngrams. default is bigram. ##train java -cp out/production/nlp:json-simple-1.1.jar LanguageDetector train nlp/model nlp/TrainData ##test java -cp out/production/nlp:json-simple-1.1.jar LanguageDetector test nlp/model nlp/TestData 2