Akademik Veri Yönetim Sistemi
Mühendislik Bilimleri ve Tasarım Dergisi, cilt.7, sa.3, ss.539-551, 2019 (Hakemli Dergi)
Named entity recognition (NER) is an information extraction (IE) task that is in the scope of natural language processing (NLP) and text mining. Its extent and methods may differbetween studies, but basically, it aims to detect expressions that indicates a person, location, organization etc. In this study, a NER structure is developed for Turkish lecture notes (for history and geography courses). Separately, this structure is a project that is specialized for an information extraction task. Besides, it also has an educational value, as the projected outcome from its execution is meaningful words or word groups from the content of input lecture notes, which can be used to construct glossary of terms structures for individual courses or course subjects. With these glossary of terms structures, it is aimed to detect expressions in the content of a lecture note that can be used for questions and support a test preparation process. In this document, general information about NER task and its scope is given; previous studies on the field are mentioned; the system developed in line with this study is introduced; success of the system is evaluated through experiment results and some thoughts for enhancement are shared.