머신러닝을 활용한 웹툰의 OSMU 가능성 예측

Abstract

The size of the domestic webtoon market is growing rapidly. The webtoon industry is a representative contents industry. Through the One Source Multi-Use (OSMU) of webtoon contents, attempts to converge with other content industries such as movies and dramas and to create new added value are gradually accelerating. Predicting webtoons with high OSMU potential can contribute to increasing the probability of successful convergence of the content industry in that digital content can be converged between multiple content industries through a single digital content. In this study, 5 machine learning based prediction models were constructed for 1,559 webtoons uploaded to Naver and Daum sites to predict the OSMU possibility of webtoons. In addition, to use webtoon images, ‘representative colors’ and ‘representative sentiment’ derived variables were created. As a result of evaluation, it was confirmed that it is possible to construct a predictive model with an accuracy of up to 72%.