An interest Mining Based Scheme for Content Dissemination in Opportunistic Networks

With the advent of the big data age, content dissemination in opportunistic networks has attracted much attention. However, problems such as sparse nodes, intermittent connectivity and limited capacity of network storage are becoming increasingly prominent, which restricts the development of content dissemination in opportunistic networks. To solve these problems, a new interest mining based scheme(IMBS) is proposed. In IMBS, the interest and meeting frequency are analyzed based on interest of mobile nodes according to Bayesian theory, to find out the human social and emotional characteristics behind the random movement of the mobile nodes. In addition, the publish/subscribe mechanism is adopted to collect the node's subscription information, in order to obtain the total demand of messages in the whole networks. When forwarding a message, the total demand of messages with the emotional characteristics and the social characteristics of the nodes to select the next-hop node are combined. Experimental results show that the proposed scheme can significantly reduce the message delivery delay and network overhead, and improve the delivery ratio of message transmission.