A Hybrid recommendation Model for CommunityAttributes of Social Networks Based on Association Rule Mining

In order to improve the joint recommendation performance of the community network, we need to design the network data crawler, and propose a community network data crawler algorithm based on association rule mining. Constructing the information transfer model of the community network, mining the information characteristic quantity of the user behavior of the community network, combining the association rules according to the attribute characteristic of the data. The fuzzy directivity clustering method is used to cluster the user behavior attributes of the community network. The autocorrelation template matching method is used to realize the information fusion and the association rules mining of the social network data. The network information crawler is implemented by combining the data clustering and distribution attributes, and the community attribute mixed recommendation is realized. The simulation results show that the proposed algorithm has higher accuracy, better personalized matching degree and higher confidence level to the community network joint recommendation results, which improves the community discovery ability.