ArmenianLanguegeAutocomplete
madanela/ArmenianLanguegeAutocomplete
armenian-nlplanguage-modelautocompletetext-generationtransformer-modelslstmweb-applicationtext-preprocessing
Summary
A project exploring Armenian language autocomplete using multiple NLP approaches including Word2Vec, LSTM, BERT translation, and GPT-2. Includes model training scripts, a web application (Flask), and data scraping/preprocessing tools. The best-performing model is a GPT-2 model fine-tuned on Armenian text.