In this project, I build a system to identify the language of written text. The text is taken from Twitter and there are nine Western European languages to choose from: English, Spanish, Portuguese, Galician, Basque, Catalan, French, Italian and German. I make a generative model, i.e., modeling p(text; language) instead of p(language|text) using a recurrent neural network (RNN) in Tensorflow.
dingluo1205 / tweet_language_detection_rnn Goto Github PK
View Code? Open in Web Editor NEWLanguage Identification of Tweets using RNN in TensorFlow
Home Page: https://dingluo1205.github.io/Tweet_Language_Detection_RNN/