FORECASTING METHOD FOR NUMBER OF AUDIENCE PER DAY OF UNOPENED MOVIE

摘要

The present invention relates to a method for estimating the number of daily audience of an unreleased movie which comprises: a pre-processing step of processing collected movie data; a group classifying step of classifying the movie data into a plurality of groups based on the number of daily audience with the processed movie data; a classification model building step of building a classification model classifying an unreleased movie into one of a plurality of groups according to the group classifying step; a regression and time-series model building step of building a regression model and a time-series model, respectively, for the plurality of groups according to the group classifying step; and an obtaining step of classifying the unreleased movie into one of the plurality of groups by using the classification model when data for the unreleased movie is inputted and obtaining time-series data for the number of audience of the unreleased movie based on the regression model and the time-series model corresponding to the classified group. According to the present invention, it is possible to effectively estimate the number of daily audience of an unreleased movie.