Research on the Association Network and Combined-Type Prediction of Films and Users Based on Complex Networks

The arrival of the era of media convergence has promoted the expansion of the film industry and the user market. At present, the data of movies and users show complex attribute characteristics, and the reasonable division of massive data is still an urgent problem to be solved in this field. Motivated by this observation, based on the classical complex network model, this article proposes the definition of object distance and the evolution rules of association network, which can be used to analyze the feature attributes of movies and users. Second, a new clustering model, in which clustering units have different interactive behavior patterns, is designed to realize dynamic clustering in association networks. Finally, we measure the market influence of different types of movies and design a prediction model of potential market user popularity of combined-types according to the related network architecture. Compared with the actual data on Douban, the rationality and accuracy of the model for market prediction of different types of combinations are verified. These findings shed new light on the practical application value of providing guidance for better film marketing production.