METHOD FOR PATTERN-ANALYZING BEHAVIOR OF AVATAR IN METAVERSE BASED ON DEEP LEARNING

摘要

The present invention relates to a metaverse avatar behavior analysis method, comprising the steps of: collecting behavior information of an avatar in a metaverse in a server and storing it in a first storage unit; performing data grouping on the behavior information of the avatar stored in the first storage unit using a recurrent neural network and storing the data in the second storage unit; analyzing the grouped data stored in the second storage unit and storing the grouped data in a third storage unit; and processing the data stored in the third storage unit and providing it as a personalized service to the avatar in the metaverse.