Intrinsic and Extrinsic Evaluation of Word Embedding Models
2020 Innovations in Intelligent Systems and Applications Conference, ASYU 2020, İstanbul, Türkiye, 15 - 17 Ekim 2020, (Tam Metin Bildiri)
- Yayın Türü: Bildiri / Tam Metin Bildiri
- Doi Numarası: 10.1109/asyu50717.2020.9259855
- Basıldığı Şehir: İstanbul
- Basıldığı Ülke: Türkiye
- Anahtar Kelimeler: extrinsic evaluation, intrinsic evaluation, word embedding, word vector, word2vec
- Boğaziçi Üniversitesi Adresli: Evet
Özet
In this study, we aimed to understand and analyze how word embedding models work on both Turkish and English. We focused on the word2vec word embedding model. We tried to improve the quality of word representations by changing the orientation of context windows. By changing the context window orientation, we aimed to train models with better accuracy results without increasing the training time. The impact of different window sizes and vector dimensions on the quality of word representations was analyzed both intrinsically and extrinsically.