Font Size: a A A

Research On Early Warning Method Of Economic Crime Based On Data Mining

Posted on:2020-06-09Degree:MasterType:Thesis
Country:ChinaCandidate:P LiuFull Text:PDF
GTID:2416330572489669Subject:Master of Engineering
Abstract/Summary:PDF Full Text Request
With the continuous development of social economy,the proportion of economic crime cases in the detection process of public security organs is relatively large and shows an upward trend.At the same time,due to the increase of factors affecting the economic crime cases involved,the empirical decision-making of economic crime case analysis is strengthened,which leads to the difficulty of early warning of economic crime and the low efficiency of prevention and detection of economic crime cases.To solve this problem,this paper proposes an early warning method for economic crimes based on data mining.Firstly,the historical data of economic crime cases are analyzed according to the early warning target,the data of economic crime are preliminarily processed,the attribute weight values affecting various economic crime cases are calculated by using the maximum correlation coefficient method and chi-square test method in the effective influencing factor screening method,the degree of influence on the attributes is ranked,the attribute values with high degree of influence on economic crime are retained,and redundant attributes are eliminated;Secondly,analyzing and comparing various data mining algorithms,aiming at the classification of economic crime data structure,selecting the decision tree method which can mine early warning rules,analyzing the defects of the traditional C4.5algorithm in the decision tree,improving the shortcomings of C4.5 algorithm in early warning rule mining,establishing an economic crime early warning model on this basis,and verifying the confidence of the model;Finally,an economic crime early warning system is established according to the economic crime early warning model,and the excavated characteristic rules are taken as early warning rules for cases under investigation,and the key persons under investigation are warned by using the characteristic matching method,so as to provide auxiliary decision-making for the daily work of the public security organs.At the same time,the reliability of the model is verified based on the sample data of a public security organ,and the efficiency of handling economic crime cases by this method is improved,which verifies that the system developed based on the economic crime early warning model has considerable practical application value.
Keywords/Search Tags:economic crime, data mining, decision tree method, early warning model, decision support
PDF Full Text Request
Related items