Türkçe için bağımlı çözümleyici kullanarak isim tamlaması çıkarımı

Noun phrase chunking is a sub-category of shallow parsing that can be used formany natural language processing tasks. In this thesis, we propose a noun phrasechunker system for Turkish texts. We use a weighted constraint dependencyparser to represent the relationship between sentence components and to determine noun phrases.The dependency parser uses a set of hand-crafted rules which can combinemorphological and semantic information for constraints. The rules are suitablefor handling complex noun phrase structures because of their flexibility. Thedeveloped dependency parser can be easily used for shallow parsing of allphrase types by changing the employed rule set.The lack of reliable human tagged datasets is a significant problem fornatural language studies about Turkish. Therefore, we constructed the first nounphrase dataset for Turkish. According to our evaluation results, our noun phrasechunker gives promising results on this dataset.The correct morphological disambiguation of words is required for thecorrectness of the dependency parser. Therefore, in this thesis, we propose ahybrid morphological disambiguation technique which combines statisticalinformation, hand-crafted grammar rules, and transformation based learningrules. We have also constructed a dataset for testing the performance of ourdisambiguation system. According to tests, the disambiguation system is highlyeffective.