Active user prediction method and device, server and storage medium

The invention discloses an active user prediction method and device, a server and a storage medium, and relates to the field of data monitoring, and the method comprises the steps: classifying each user according to user attribute information, so as to obtain the user category of each user; according to the active event set of each user category, obtaining an active influence coefficient of each user within a first preset time; according to the historical login information and the active influence coefficient of each user, through a pre-trained prediction model, obtaining predicted active time of each user in a first preset time; wherein the prediction model is constructed based on a neural network structure. According to the technical scheme of the embodiment of the invention, the accuracy of the prediction result of the active user is improved, the predicted active duration of each user is obtained, the prediction range of the active user is expanded, and the prediction precision of the active user is impro