Chinese ceramic culture has a long history, and Chinese ceramics are closely related to the process of Chinese history. In each era, the dissemination and evolution of Chinese ceramics has formed its own characteristics in the dissemination background and channels, dissemination characteristics, and dissemination influences. Ceramic culture is the common artistic language of mankind, which exists in the form of a special cultural symbol. It has a long history and profound cultural heritage, and plays an important role in Sino-foreign exchanges. However, there are many difficulties and challenges in the process of communication and dissemination with cultures from all over the world. It is increasingly an important topic to study the propagation path of ceramic culture, so as to promote overseas spread for ceramic culture. This work uses complex networks to predict and analyze the propagation path of ceramic culture. First, this work selects the BP network as the basic framework to predict the propagation path of ceramic culture. Neural networks are complex networks that can perform such tasks efficiently. Second, because the BP algorithm has problems such as slow convergence and easy to be affected by the initial weight, this work uses the PSO algorithm to make up for the shortcomings of BP algorithm. To solve slow convergence in the later stage of the PSO algorithm and easy to fall into local optimum, this paper adopts improved strategy with dynamic inertia weight and dynamic learning factor. Third, this work integrates the improved PSO (IPSO) with the BP network to construct an IPSO-BP network, and realizes the prediction of the propagation path of ceramic culture based on IPSO-BP.