Research On Data-Mining Oriented Criminal Investigation

The public security has become more and more complicated and severe, so the police have to face with greater pressure to criminal investigation. With the development of the Golden Shield Project, the application of data mining technology on investigation will further improve the police's work efficiency extensively. Respecting the raw data of application on investigation which have high distribution in space and apparent dynamics in time, this paper mainly focuses on the data mining which applys on the investigation.Firstly, with the analysing and comparing of the fundamental theories and approaches of data mining and data warehouse, this paper improves the traditional model of data mining in the reserve way and gives the model of dynamic data mining with multi-data source and the system structure, the new model has a better characteristics than the traditional model of data mining. Secondly, based on the characteristics of data warehouse sources of investigation and introduction of data exchanging platform, a data warehouse of investigation is designed and implemented. Finally, since the traditional Apriori algorithm has some deficiencies, two Aprori optimization algorithms based on the new crime sensitivity parameter and the priority parameter are presented respectively, also the feasibility and effectiveness of two Aprori optimization algorithms are validated by simulation.