A Comparison of Word Embedding Models for Turkish

A comprehensive analysis of static word embeddings for Turkish

“…This method can achieve state-of-the-art performance even without employing language-dependent features for morphologically rich languages such as Turkish and Czech. (Dündar & Alpaydın, 2019) A recent study conducted by (Bayrak et al, 2023) involved a comparison of Turkish word embeddings trained using Word2Vec, FastText, ELMo, and BERT models. These embeddings were input into an LSTM text classification model.…”