Font Size: a A A

Research On Model Of Association Analysis For Criminal Cases Of City Police

Posted on:2008-02-19Degree:MasterType:Thesis
Country:ChinaCandidate:X J ZhangFull Text:PDF
GTID:2166360218455606Subject:Information management and e-government
Abstract/Summary:PDF Full Text Request
The information project of police in our country, JinDun project, comes into a new stage. We have got tremendous achievements, but there are still some problems. One is that the all kinds of information resource has not been fully mined and used and can't give the information support and decision support for the leader and front-line police by simple and flexible means. Recently, the decision-maker can get more information from the hidden rules found by data mining technology. So, the model of association analysis for criminal cases which is one of the most important pans in data mining for city police is built in this paper.First, the meaning of research is analyzed in the paper and the application that data mining applied in the police system in all over the world is reviewed. Meanwhile, the paper introduced the basic definition and arithmetic of data mining and association analysis. Then, this paper proposed the main frame of association analysis of criminal cases for city police and explained the function, method and arithmetic in every part. There are three main pans. Data pre processing: when the method of data extraction is proposed, data cleaning is the next step according to the concrete data. On the basis of proper data, the star schema is chosen, data discretization and concept hierarchy is completed by binning and the data cube is built at last. Mul-dimension association rules mining: after built the data model, this paper proposed the arithmetic of frequent verb set improved on Apriori. According to the minsup and mincof, the strong association rules were gained. Mul- hierarchy association rules mining: when mining the mul-dimension association rules, the association rules can be found in different concept hierarchy using the strategy of decreasing support. The frequent verb set was found by developed arithmetic in mul-dimension mining and strong rules come out by the same way. Especially, there were redundant results. So, the last step is reducing the meaningless results. At last, according to the method and process of model, the model is checked using the real data from DaLian police by proper tools. The model of association analysis for city police is carried out. It proved the validity and effectiveness of model.The model built by this paper will paly a significant role in the data mining application reseach for city police.
Keywords/Search Tags:Criminal case, Concept hierarchy, Data cube, Mul-dimension rules, Mul-hierarchy rules
PDF Full Text Request
Related items