Prediction of Future Possible Offender’s Network and Role of Offenders

Abstract

Data mining is a process of extracting information and discovering patterns from large amount of data. Crime Analysis can be a lucrative application of data mining. Data mining can play an important role in analyzing and predicting crimes using the data stored in repositories. Crime rates all over the globe are increasing day by day, has became a topic of major concern. Due to awful growth in crime rate, it has become impossible to analyze those crime related data and detect crime patterns or predict future crimes by intelligence agencies or local law enforcement agencies manually. Using data mining techniques we can produce important results like, detecting crime patterns and relationship among criminals. Solving crimes is a difficult and complex task that requires human intelligence and experience and data mining can be very useful for the peoples involved in solving crime. Methods such as clustering, classification, link analysis, graph based approach etc. can be used to analyze crime patterns or they can be used to predict future crimes. Nowadays, there are a number of algorithms like, J48, Naïve Bayes, K-means, Apriori etc., related to data mining are available which we can implement in our crime investigation. All these analysis will be helpful in making strategy and tactics to address crime and disorder. Also, in order to discover those offenders that are involved in a conspiracy, drug trafficking, terrorist activities, etc. together, we need to apply mining on co-offenders network. In this paper we are going to study a crime data model which can efficiently analyze the co-offenders network to provide useful information about offender’s activity and groups. Information like, formation or dissolution of criminal group, hidden links between offenders, role of offender, identifying central actors etc. can be derived using co-offender network mining which can be used by law enforcing agencies to make strategies and deciding further steps to stop crimes and catching crucial criminals This thesis presents a detailed analysis of co-offender’s network, the collaboration and dissolution of crime groups, extraction of various relevant crime patterns, hidden links, link prediction and centrality analysis of crime data in order to uncover the role of offender. This study will be helpful to law enforcing agencies in making strategies and tactics to address crime and disorder.