Idioms as multi-word expressions in Turkish


Thesis Type: Postgraduate

Institution Of The Thesis: Middle East Technical University, Turkey

Approval Date: 2020

Student: Arzu Burcu Güven

Supervisor: HÜSEYİN CEM BOZŞAHİN

Abstract:

Idioms constitute several challenges for both Natural Language Processing (NLP) and linguistic analysis. A better understanding of idioms will yield valuable insights about natural language as well as the way it is processed. The relevance of idioms, along with the fact that Turkish is a rather unexplored language from this perspective, motivates us to work on Turkish idioms. Here, we aim to demonstrate a grammatical study on Turkish idioms that were selected in accordance with distributional models.