편년체 자료의 통계 정보 추출과 활용 –『日省錄』·『承政院日記』·『朝鮮王朝實錄』을 중심으로

Abstract

Until now, humanities-related search systems have provided services centered around original images or texts, as well as interpretation information. This type of search-focused service has certainly helped humanities research and contributed to the development of humanities. On the other hand, Ilseongrok(日省錄, the Daily Records of the Royal Secretariat), Seungjeongwon Ilgi(承政院日記, Diaries of the Royal Secretariat) and the Joseon Wangjo Silrok(朝鮮王朝實錄, Veritable Records of the Joseon Dynasty) are valuable chronological materials that have been listed as UNESCO World Documentary Heritage, each containing historical records of political, economic, social, cultural and meteorological events spanning from 150 to 500 years. Despite these, statistics on these documents have not been published, and there is a need for research to effectively analyze them. In the past, research have been conducted in the form of “original text search → confirmation of specific facts → (statistics extraction through researchers’ hard work) → conclusion”. However, if the thinking is reversed to “statistics extraction → confirmation of specific facts → original text search → conclusion”, it is believed that vast records can be effectively utilized. To this end, the amount of data and the characteristics of the data were analyzed first to find a way to visually represent the search terms entered by the user. This paper suggests that statistical information can be extracted using tagged information, such as people and places, provided by the data search system, and that weather information and the frequency of search terms can be extracted by year, month, and day, and presented in Google charts’ line graph. It also highlights that the relationship between the generation and disappearance of search terms can be found over time, and that historical facts can be effectively verified. In addition, it provides convenience by integrating searches for the three scattered and serviced records, allowing for a simultaneous comparison of original texts for the same date. This research is one of the methods of studying statistical-centered humanities knowledge visualization modeling, which is advocated by the Chinese Character Culture Institute at Chungnam National University as part of a preliminary study for classifying personal collections included in the “Korean Collection of Books” published by the Korean Classics Translation Institute. This is an example of finding the problems of search-focused services provided to humanities researchers and suggesting a direction for humanities-related software to move forward.