Volume 29, Issue 4, 2020
DOI: 10.24205/03276716.2020.868
Research on Early Warning Methods of Economic Crimes Based on Data Mining
Abstract
This paper presents an early warning method of economic crime based on data mining. First, analyze the historical data of economic crime cases for the early warning targets, perform preliminary data processing on economic crime data, and use the maximum information coefficient method and the chi-square test method in the effective influencing factor screening method to calculate the attribute weights that affect various economic crime cases. Value, ranks the degree of influence of the attribute, retains the attribute value with high degree of influence on economic crime, and removes redundant attributes. Secondly, analyze and compare a variety of data mining algorithms. According to the characteristics of the economic crime data structure, the decision tree method that can mine the early warning rules is selected, and the defects of the traditional C4.5 algorithm in the decision tree are analyzed. The deficiencies in the mining of early warning rules are improved, and an economic crime early warning model is established on this basis, and the confidence of the model is verified: Finally, an economic crime early warning system is established based on the economic crime early warning model, and the excavated characteristic laws are used as the investigation The early warning rules of cases use feature matching methods to warn suspects and high-risk followers under investigation to provide auxiliary decision-making for the daily work of public security organs. At the same time, the reliability of the model is verified based on the sample data of a public security agency, and the efficiency of using this method to deal with economic crime cases is improved, which verifies that the system developed based on the economic crime early warning model has considerable practical application value.
Keywords
Economic crime, data mining, decision tree method, early warning model, auxiliary decision