A Pilot Study on Automatic Inference Rule Discovery from Turkish Text

Isguder-Sahin G. G., Adali E.

8th IEEE International Conference on Application of Information and Communication Technologies (AICT), Astana, Kazakistan, 15 - 17 Ekim 2014, ss.268-272 identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Basıldığı Şehir: Astana
  • Basıldığı Ülke: Kazakistan
  • Sayfa Sayıları: ss.268-272
  • İstanbul Teknik Üniversitesi Adresli: Evet

Özet

In this paper we present a framework for extraction of inference rules from Turkish documents, such as "A solved B similar to A found a solution to B". Many natural language processing tasks, such as question answering, information retrieval and machine translation can benefit tremendously from inference rules. Our framework consists of three layers: construction of dependency trees; determining predicates and their complements and finally discovery of inference rules via clustering. We use Silhouette Index (SI) of the clusters as an automatic evaluation metric.