빅데이터를 활용한 섬 경관 특성 분석 - 전라남·북도 섬을 대상으로 -

Abstract

This study attempted to use big data analysis as a way to supplement the limitations of existing landscape research while increasing the need to identify island landscape characteristics, such as indiscriminate island development and projects that do not consider island characteristics. Therefore, the purpose of this study was to present basic data on plans and management measures considering the unique island landscape characteristics, which are important indicators of tourism motivation. The scope of the study was limited to Cheongsan-do, Seonyu-do , Bogil-do, and Bakji ·Banwol-do Island, which have easy data for big data analysis among islands recognized at home and abroad. The collected data are photo and text data, and the results were derived by classifying them into formal aesthetic and psychological landscape characteristics in order to identify each landscape characteristic and analyze the relationship. For formal aesthetic landscape characteristics, landscape models, colors, and color image adjectives were analyzed based on photo data to identify landscape components and characteristics of each island, and psychological landscape characteristics were analyzed by frequency analysis, N-gram, network analysis, and emotion analysis using text data. By synthesizing the above results, the meaning of how the island landscape is used by the public was grasped, and the existing research could be supplemented in that it could grasp the temporal characteristics such as seasons and time series as well as the limitations of time and cost incurred in island research. In addition, it suggested that it could be a device that can further specify the direction of island development in the future in consideration of the landscape characteristics of each island.