??????? ?? ??? ?? - <????> ?? ????? ?? ??? ???? - (A Study on the Place-name for Natural Language Processing: focusing on the Classification of Types of Urimalseam)

? ??? ??????? ??? ???? ?? ??? ???? ?? ???? ??? ??? ?? ? ??? ??. ??? ??? ??? ??? ??? <????>? ��??�� ???? ?? ? ??? ?? ?? ??? 13,671?? ???? ???. ???? ?????? ???? ?? ???? ?? ???? ??????? ??? ? ??? ?? ???? ?? ?????.
??????? ??? ?? ??? ?? ��TTA ??? ?? ??? ?????��? ???? ??? 15? ??? ??(LC, Location)? ???? ???. ??? ???? ?? ???? ??? ?? ??? ???? ??? ???? ???? ?????? ?????. ?? ????? ?, ??, ?, ?? ??? ??? ???? ????, ???? ??? ??? ??, ??, ???, ??? ??? ???? ???. ??? ???? ???? ??? ??? ?? ???? ??? ? ??? ???.

The purpose of this study is to classify the types of place-name for natural language processing and to examine the characteristics of the place-name in Korean. For this purpose, 13,671 place-name in the Korean Peninsula were used in the previously published user-centered dictionary, Urimalsam. It was classified based on the rear part of the place-name, and the target was classified in detail to be used for natural language processing.
For the classification of types suitable for natural language processing, reference was made to the TTA named entity tag set and guidelines, followed by the class of place-names(LC, Location) among 15 sub-tag set. In this process, the types were classified according to the distribution of the place-names in the Korean peninsula and the rear part of place-names were presented. The Korean place-name has a variety of vocabulary related to rivers, seas, mountains, and bay, and the types of place-name of scenic spots were also divided into streets, parks, tourist attractions, and amusement parks. It is stated that even the same designation can be tagged into different classes depending on the context.