Community detection in an integrated Internet of Things and social network architecture

In this paper, we propose a community detection scheme in an integrated Internet of Things (IoT) and Social Network (SN) architecture. The paper takes a graph mining approach to solve the problem in complex network of IoT and SN. A number of pieces of research literature exist on community detection in SNs; however, no work specifically on integrated IoT and SN architecture addresses this issue. The existing community detection approaches have not considered things into account. We propose the scheme, Community Detection in an Integrated IoT and SN (CDIISN) in which we divide the nodes/actors in complex networks into basic nodes and IoT nodes, and execute the community detection algorithm. We consider two nodes to be in a community, only if the nodes are at most one hop apart and have at least two mutual friends. The smallest community in our case is a subgraph with a cycle of length four. In our approach, a node can be part of multiple communities, and it works well for weighted graphs. Once communities are extracted, we use an access control scheme, based on which access to nodes is provided. This approach of community detection in an integrated environment would find tremendous use in the future, because in the case of any search operation performed by any node, the results obtained intra-community are more relevant than inter-community.