A Turkish language factoid question answering system based on Deep Learning Neural Networks
Is this your thesis?
This record came from a bulk archive import. If it’s yours, link it to your profile.
Abstract (EN)
The aim of this thesis is to get a meaningful answer to a question based on information taken from a set of documents where the complete answer can be found inside a text. A factoid question answering system allows to answer questions related only with facts (e.g. which is the capital of Turkey). The output will be a closed-domain Turkish factoid Question Answering (QA) system that is intended to be used by customers and employees as an assistant in a Turkish Bank. The documents are unstructured and written in Turkish language. Deep Learning algorithms will be used over the data such as BERT.
Author
Cavide Balkı Gemirter
Institution
How to Cite
Cavide Balkı Gemirter (Master Thesis). A Turkish language factoid question answering system based on Deep Learning Neural Networks, 2019, Yeditepe University.
Keywords
License
Tüm Hakları Saklıdır
This work is shared under the specified license terms.
More theses from Yeditepe University
- Studies on cyclodextrin complexation of a poorly water soluble anti-hyperlipidemic drug, tablet formulation and characterization(2021)
- Washington ambassadors in Turkish-US relations (1927-1960)(2023)
- Metamorphosis of female voices: A study of the violation of women in Greek and Roman mythology and feminist rewritings reclaiming the narrative(2022)
- Knowledge distillation with foundation models for image segmentation(2023)
- The relationship between machiavelism, grandiose and vulnerable narcissism, and loneliness among white collar workers(2023)
- Evaluation of drug-drug interaction checkers along clinically relevant adverse drug events in oncology and hematology pediatric patients(2023)