논문 상세보기
Exploring Nuclear Future Signals Using Scientific Article Data
The purpose of this study is to detect future signals and changes in nuclear-related research to apply safeguards by design to new nuclear facilities or to determine nuclear fuel cycle-related research and development (R&D) activities. First, a total of 2,029 scientific articles published between 2015 and 2022 in the journal of “Nuclear Engineering and Technology” by the Korean Nuclear Society were collected. The authors of the scientific article used their expertise and knowledge to select keywords that can properly represent the article. Therefore, in this study, the keywords of each scientific article were analyzed using the technique of text mining. We then calculated the “word frequency” and “term frequency-inverse document (TF-IDF)” values of the keywords. Consequently, significant words such as “reactor,” “nuclear,” and “fuel” were extracted, which were represented as word clouds. Furthermore, keywords extracted through text mining were quantitatively classified into weak or strong signals using a keyword emergence map (KEM). The KEM is a tool that can explore future signals because essential keywords have a high frequency of appearance, and newer keywords are more important than older keywords. The KEM results showed no keywords in the strong-signal area in the field of nuclear academia. However, keywords such as “deep learning,” “earthquake,” “zircaloy,” and “CFD” were confirmed to be distributed in the weak signal area. A weak signal indicates the most probable topic that could become a strong signal in the near future. The weak signal methodology can be applied to predict future nuclear scientific trends in the rapidly changing world. Based on the results of the study, changes in the subject of nuclear-related scientific articles over the past eight years and future signals were interpreted. The results confirmed that this method can be applied to safeguards measures of new nuclear facilities in the design stage and can be used to detect R&D activities related to the nuclear fuel cycle in advance.
