Cangqi Zhou

TL;DR: A two-stage method for detecting community structure in bipartite networks is proposed, developing the Bi-Louvain algorithm that iteratively groups the nodes in each part by turns and demonstrating that the calculation of the gain of modularity of each aggregation, and the operation of joining two communities can be compactly calculated by matrix operations for all pairs of communities simultaneously.

TL;DR: A parsimonious model is proposed to predict the saturated numbers of forwarding activities of messages, which reflects average people’s intrinsic psychological gain under repeated stimuli and supports that there exists a relatively fixed gain brought by repeated exposures.

TL;DR: A new recommendation method based on collaborative filtering called User-Item Community Detection based Recommendation (UICDR) method, modified from the previous work, that significantly improves the performances of Top-N recommendations of several traditional collaborative filtering methods.

TL;DR: A discrete-time Markov chain method is proposed to model susceptible-infected-susceptible epidemic dynamics in heterogeneous networks and finds that the dominations of different types of residences might reverse when mobility probability varies for some networks.

TL;DR: This paper proposes an intuitive yet effective graph convolution module to aggregate local attribute information of a given node to obtain rational node feature representations and builds a label-aware representation learning framework to measure the compatibility between pairs of node embeddings and label embedDings.