With the development of social media, social networks have become an important platform for people to share and communicate. In social network communication, each participant can only pass information correctly if we find the goal of our communication. In other words, when people carry mobile devices for data transmission, they need to find a definite transmission destination to ensure the normal conduct of information exchange activities. In social networks, this is manifested in the process of data transmission by nodes, which requires analysis and judgment of surrounding areas, and finds suitable nodes for effective data classification and transmission. However, the node cache space in social networks is limited, and the process of waiting for the target node will cause end-to-end transmission delay. In order to improve such a transmission environment, this paper proposes a node trajectory prediction method named EDPPM algorithm. This algorithm can guarantee that nodes with high probability are given priority to obtain data information, which realized an effective data transmission mechanism. Through experiments and comparison of opportunistic transmission algorithms in social networks, such as Epidemic algorithm, Spray and Wait algorithm, and PRoPHET algorithm, our algorithm can improve the cache utilization of nodes, reduce data transmission delay, and improve the overall network efficiency.