Effect of the training set on the word embeddings and similarity test set for Turkish

• Published in: 2016 24th Signal Processing and Communication Application Conference (SIU) pp. 1005 - 1008
• Main Authors: Yucesoy, Veysel, Koc, Aykut
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.05.2016
• Subjects: Blogs; classification; Internet; Natural language processing; Postal services; Semantics; Silicon; Training; Turkish similarity test set; Word embeddings

Word embedding, which is usually used in the literature especially for English, is a technique to associate each word to a mathematical vector representation under which some structural or semantic relations hold. There are some Turkish application of this technique. Despite being designed according...