Yazar "Esgel, Volkan" seçeneğine göre listele

Listeleniyor 1 - 2 / 2
  • Küçük Resim
    Yayın
    Construction of a Turkish proposition bank
    (Tubitak Scientific & Technical Research Council Turkey, 2018) Ak, Koray; Toprak, Cansu; Esgel, Volkan; Yıldız, Olcay Taner
    This paper describes our approach to developing the Turkish PropBank by adopting the semantic role-labeling guidelines of the original PropBank and using the translation of the English Penn-TreeBank as a resource. We discuss the semantic annotation process of the PropBank and language-specific cases for Turkish, the tools we have developed for annotation, and quality control for multiuser annotation. In the current phase of the project, more than 9500 sentences are semantically analyzed and predicate-argument information is extracted for 1330 verbs and 1914 verb senses. Our plan is to annotate 17,000 sentences by the end of 2017.
  • Küçük Resim
    Yayın
    Semantic role labeling for Turkish propbank
    (Işık Üniversitesi, 2019-09-06) Esgel, Volkan; Yıldız, Olcay Taner; Işık Üniversitesi, Fen Bilimleri Enstitüsü, Bilgisayar Mühendisliği Yüksek Lisans Programı
    People's communication with each other takes place through sentences that combine words with different purposes. Words can gain different meanings with the presence of other words in the sentences in which they take place. With the rapid development of technology, the studies on understanding of human language by computational power have gained speed. These studies are generally referred to Natural Language Processing and their main purpose is to understand the sentences in human communication. The words in the sentence ful l different purposes. Some words describe an event, while other words indicate details of that event. De ning the semantic roles of words is possible with different algorithms. This study rst started by contributing to the process of determining the semantic roles of the word groups in the sentence by manpower. In addition, the semantic roles in the English sentences were parsed and shared on a web site with the marked roles in the Turkish sentences for comparison purposes. Finally, it is tried to measure how the algorithms aiming to nd the semantic roles of the words in the sentence perform automatically for Turkish.