초록 열기/닫기 버튼
In order to quickly respond to changes in marine tourism trends and to establish effective policies, it is necessary to prepare a reliable method for collecting data on visits. The purpose of this study is to conduct an exploratory study on the behavior of marine leisure tourism users by using images, which are unstructured data of public data and social media data, in order to analyze the characteristics of domestic marine tourism. As a research method, the possibility was proposed as a new research method by using the convolutional neural network(CNN) algorithm used for image classification based on deep learning. Supervised learning was performed by classifying behaviors based on theories based on public data and images collected through SNS. As a result of the study, the behaviors of beach users were largely classified into landscape-based and activity-based. As the supervised learning of this study did not show satisfactory results, it is necessary to classify marine leisure tourism behavior by applying the photo classification algorithm for future research. In addition, a management plan for public data was proposed for continuous research use.
키워드열기/닫기 버튼
Marine Tourism, Beach, Image Classification, Deep-learning, Convolutional Neural Network
