抽象的な

Detecting crime types using classification algorithms

Cuicui Sun, Chunlong Yao, Xu Li, Xiaoqiang Yu

Criminal behavior reflects the characteristics of the criminals. To infer the types of unknown criminals from vast amounts of different crime characteristics is an important part of criminal behavior analysis. It is a good solution to classify the criminals using classification algorithms. Three typical classification algorithms are used to analyze the criminal datasets in this paper, including C4.5 algorithm, Naive Bayesian algorithm and K nearest neighbor (KNN) algorithm. However, quite a lot of missing data values can result in a seriously effect on the classification accuracy. Therefore, the missing data filling method which fills missing data based on grey relational analysis (GRA) theory is used. The experimental results on the criminal dataset show that higher classification accuracy can be obtained using this missing data filling method