Social network analysis for crime rate detection using Spizella swarm optimization based BiLSTM classifier

Abstract

Technical advancements initiated a rapid increase in criminal activity over time. For the prevention of these criminal activities, preventive measures are needed. In order to monitor these illicit activities and enhance public safety, crime rate detection is essential. Social media can be used to identify crime rates in various parts of any nation, which can dramatically lower crime rates. Social media are a source of information as well as a tool for communication. Twitter, which has a user base of more than 300 million, makes a suitable choice for data analysis. The Spizella swarm based BiLSTM classifier is used for the detection of crime rate in this research. While performing social network analysis using the BiLSTM classifier for the determination of crime rate providing faster convergence is a crucial factor and this faster convergence is achieved by the proposed Spizella swarm optimization. BiLSTM classifier effectively identified the crime rate and the BiLSTM performance is boosted by the Spizella swarm optimization where the escaping characteristics of Spizella improve the convergence and help in attaining desired results Additional training is given by the BiLSTM classifier by traversing the outputs and this BiLSTM classifier are more efficient in the text classification. Measuring the metrics values for accuracy, sensitivity, and specificity demonstrates the effectiveness of the proposed method and the proposed Spizella swarm optimization achieved an improvement of 0.5 %, 1.16 %, and 1.08 %, which is more efficient.