Generalized Jaccard Similarity Based Recurrent DNN for Virtualizing Social Network Communities

In social data analytics, Virtual Community (VC) detection is a primary challenge in discovering user relationships and enhancing recommendations. VC formation used for personal interaction between communities. But the usual methods didn’t find Suspicious Behaviour (SB) needed to make VC. The Generalized Jaccard Behavior Similarity-based Recurrent Deep Neural Network Classification Ranking (GJSBS-RDNNCR) Model addresses these issues. GJSBS-RDNNCR model comprises four layers Social Networks (SN). model, SN given as an input at layer. After that, User’s Behaviors (UB) are extracted first Hidden Layer (HL), Similarity coefficient calculates similarity value second HL based on SB. third HL, values examined, SB tendency classified using Activation Function (AF) Output (OL). Finally, ranking process performed with users their Results analysis metrics such Accuracy (CA), Time Complexity (TC), False Positive Rate (FPR). experimental setup considers 250 tweet from dataset identify SBs of users.