Automatic Movie Tag Generation System for Improving the Recommendation System

As the content industry develops, demand for movie is increasing. Accordingly, actively developing super-personalized recommendation systems that match consumers’ tastes. In this paper, we study automatic generation of tags to improve system. We extracted background sounds from trailer videos, analyzed using STFT (Short-Time Fourier transform) and major audio attribute features, created a genre prediction model. The experimental results show pre-collected dataset data via model are similar when compared. research, suggest methodology an system information videos. This will help reduce time effort metadata