Traditional visual communication design teaching courses only use books and courseware for teaching, which greatly limits students understanding and perception of visual communication design art courses. The visual communication design course is different from the teaching of other subjects; it requires students to have a deep understanding of the content of the video and the experience of emotion. Relying on courseware and book teaching methods will not only limit students’ imagination, but also make students tired of learning psychological activities. The teaching of visual communication design technology is to transmit the images, colors, words, and emotions of artworks to students in the form of video, which is a relatively abstract course. With the development of big data technology, this brings new opportunities for the teaching of visual communication design technology courses. In this study, data mining technology will be used to evaluate the effect of visual communication design teaching. And CNN will be used to predict the content characteristics of visual communication design teaching, which is part of the visual communication design teaching system. The research results show that the big data method has better performance in the visual communication design technology course; both the classification and prediction errors are within the acceptable range for the artwork of the visual communication design course. Big data technology can well predict the relevant features in visual communication design; the largest prediction error is only 2.66%, and the smallest error is only 1.21%.