Crime Analysis using DBSCAN Algorithm

Devvrat Mungekar, Himani Joshi, Adinath Kankekar, Pratap Nair, Poulami Das

20212021 Third International Conference on Inventive Research in Computing Applications (ICIRCA)15 citationsDOI

Abstract

Today with the growth in the human population, the crime rate has also escalated. To mitigate the increasing crime rate, it is necessary to analyze this humongous data using data science and machine learning techniques. The existence of a system that makes it easier to analyze, visualize and predict criminal data is indispensable. In this paper a user-friendly web application has been proposed that investigates criminal data using data analysis and data visualization methods. Using this system, the user can understand the crime patterns, crime trends, identify the criminal hotspots by observing the results from the data visualization methods. Density based Spatial Clustering of Applications with Noise (DBSCAN) algorithms has been used to cluster the crimes according to the location. The Crime Analysis Web Application is developed using the open-source framework Streamlit and Google Cloud Platform's (GCP) BigQuery engine.

Topics & Concepts

DBSCANComputer scienceVisualizationCrime analysisCluster analysisCloud computingData miningData visualizationPopulationData modelingNoise (video)Data scienceMachine learningDatabaseArtificial intelligenceImage (mathematics)Canopy clustering algorithmCorrelation clusteringSociologyCriminologyOperating systemDemographyTraffic Prediction and Management TechniquesAnomaly Detection Techniques and ApplicationsVideo Surveillance and Tracking Methods