Digital media is an inevitable product of the development of information technology, and the emergence of this art has brought new opportunities and space for the innovative development of the film industry. This paper constructs a film creation model based on digital media technology and analyzes its data set using data mining algorithms. Firstly, the digital media is set as a matrix to calculate the similarity, and a subset of attributes is calculated using cosine similarity and Pearson coefficient to extract the indistinguishable relationship of rough set theory. Then the information entropy calculation scheme is reset, and the specific process of the data mining method is given. Finally, the mining is carried out for six data subsets, from which the importance of the integration of digital media and film creation is known. The results show that the sound and picture fluency tests of the frames in the film creation using digital media technology reached 8 frequencies for all three kinds of audio during the sound test. The picture smoothness of each frame of the film creation work met the design standard of >16.67ms/frame. Therefore, the combination of digital media art and film creation in modern film creation can not only promote the development of the film industry but also enhance the aesthetic ability of the audience to a certain extent.