Abstract:
Since the birth of civilization, offenses of various sorts have been on an uptrend and no wonder Crime investigation, which is emerging as the supreme law enforcement procedure , particularly of the Government, to check such menaces to the society at large, finds itself engrossed in methodical scrutinizes for spotting and evaluating the designs and tendencies shown by the law-breaking anti-socials indulging in offense and chaos so as to find appropriate and accurate preventive actions in due time. Various Techniques were introduced for the method of Crime analysis and Prevention. But the existing methods have some drawbacks, i.e. those methods are not considering the precise features to analyze and predict the high volume crime areas. Hence to reduce the drawbacks in the existing methods, a new crime location prediction technique is proposed in this paper. The proposed technique predicts the crime location by analyzing the crime data by utilizing an Adaptive Mutation based Artificial Bee Colony (AMABC) algorithm. The AMABC algorithm will use socio-economic factors and clustering results in the crime location analysis process. Among the predicted crime locations by the AMABC algorithm, a high crime location is computed by mining the patterns by using Association Rule Mining (ARM) technique. Thus, the proposed technique will successfully predict the locations via AMABC and ARM techniques. In our proposed technique an UCI Machine Learning Repository-Communities and Crime Data Set will be used for the crime analysis. The proposed technique will be compared with the existing optimization methods like GA, PSO and conventional ABC.
Page(s):
45-56
DOI:
DOI not available
Published:
Journal: Journal of Theoretical and Applied Information Technology, Volume: 59, Issue: 1, Year: 2014