Generating a Concept Relation Network for Turkish Based on ConceptNet Using Translational Methods
22nd International Conference on Speech and Computer, SPECOM 2020, St. Petersburg, Rusya, 7 - 09 Ekim 2020, cilt.12335 LNAI, ss.369-378, (Tam Metin Bildiri)
- Yayın Türü: Bildiri / Tam Metin Bildiri
- Cilt numarası: 12335 LNAI
- Doi Numarası: 10.1007/978-3-030-60276-5_36
- Basıldığı Şehir: St. Petersburg
- Basıldığı Ülke: Rusya
- Sayfa Sayıları: ss.369-378
- Anahtar Kelimeler: Common sense knowledge base, ConceptNet, Turkish, Word sense disambiguation
- Boğaziçi Üniversitesi Adresli: Evet
Özet
ConceptNet is a large-scale network of concepts and relationships, based on various common sense knowledge bases. Turkish is a language that lacks similar resources for processing texts and extracting meaning. This study discusses various methods to create a Turkish ConceptNet using translational techniques based on English ConceptNet and explains the results herewith obtained. Multiple models were tested, using different knowledge sources and tools including WordNet, Wikipedia, and Google Translate. Results obtained from each model and the approaches to improve these results are discussed, while also explaining details, assumptions, and drawbacks relevant to each relation.