In the big data environment, the social information of a large number of nodes cannot be reasonably analyzed and utilized, thus leading to the problem of uneven routing performance. Therefore, this paper proposes a Social Similarity Routing Algorithm (SRRA) based on socially aware networks in the big data environment. In the SRRA algorithm, two main parts in which the process of nodes forwarding messages are in-community and out-of-community. First, we defined three indexes for nodes and communities in which nodes are located by analyzing human social behavior: community connectedness between communities, the activity of nodes, and the social similarity of nodes. Then these three indexes are used to make up two measures: the in-community forwarding measure and the out-of-community forwarding measure. When messages are forwarded within a community, we choose nodes with high in-community forwarding measures as relay nodes so that messages can be delivered quickly in the same community. The relay node with the highest out-of-community forwarding measure is chosen to forward the message to the adjacent communities that are as near as possible to the destination community as much as possible when messages are forwarded outside the community, which ensures that messages can always be sent to the target community fast and accurately. The results of the simulation experiments compared with existing routing algorithms prove that the SRRA routing algorithm significantly improves the message delivery ratio while effectively reducing the network overhead and average latency.

在大数据环境下，大量节点的社交信息无法得到合理的分析和利用，从而导致路由性能参差不齐的问题。因此，本文提出了一种社会相似度路由算法（SRRA）基于大数据环境中的社会意识网络。在SRRA算法中，节点转发消息的过程主要分为社区内和社区外两个部分。首先，我们通过对人类社会行为的分析，定义了节点和节点所在社区的三个指标：社区间的社区连通性、节点的活跃度和节点的社会相似性。然后用这三个指标组成两个度量：社区内转发度量和社区外转发度量。当消息在社区内转发时，我们选择社区内转发量大的节点作为中继节点，以便消息在同一社区内快速传递。选择社区外转发量最大的中继节点，在消息转发到社区外时，将消息转发到尽可能靠近目标社区的相邻社区，保证消息可以通过始终快速准确地发送到目标社区。仿真实验结果与现有路由算法对比证明，SRRA路由算法在有效降低网络开销和平均时延的同时，显着提高了消息传递率。