Research On The Discovery Method Of Suspicion Degree Relationship Based On The Association Rule Algorithm

In recent years, the construction of public security information has made a rapid progress and accumulated vast amounts of data. However, the use of these data is limited to traditional approaches now, such as statistics, query, update.The researches and applications of these data are still in their infancy. The hidden rules and clues in crime data are analyzed by data mining, which needs to be further studied. The association analysis for the suspects’ characteristics can mine criminal rules and clues implied in crime data and make a basis for the following criminal investigation.The interesting rules or association can be found from a large amount of data by association rules, which offers help for analysis and decision later. Now in the analysis of the suspects’ characteristics, folks mostly use the traditional Apriori algorithm to do the work.But when analysising a large amounts of data, Apriori algorithm exists a problem of low efficiency. So there are many improved algorithms for the shortage of Apriori algorithm. The MC_Apriori algorithm based on compressed matrix showes good performance.However, for the problems in the redundancy of the data, there is no solution in the algorithm.It will have a great influence on the efficiency of the algorithm.Based on MC_Apriori algorithm, this paper studies the redundant data and puts forward an ICM_Apriori algorithm. The algorithm adds a row and tow columns,which compresses the matrix by dealing with redundant items, the unnecessary itemsets which can’t be connected as well as the infrequent ones,which Then through a lot of experimental data, ICM_Apriori algorithm is verifed. Based on the analysis of others algorithm, it proves that ICM_Apriori algorithm has very good effectiveness.Finally ICM_Apriori algorithm based on the compression matrix is applied to the association analysis for the suspects’ characteristics to analysis information of criminals, such as the case category, gender, age, residence, height, length of foot, literacy and marital status. It gets the relevant criminal laws and rules and converts to the information of early warning mechanism and intelligence to make technical support and basis for forensic work.