An integrated Internet geographic information system for crime control | |
| Author | Roongrasamee Boondao |
| Call Number | AIT Diss. no.IM-06-01 |
| Subject(s) | Crime prevention--Geographic information systems Internet |
| Note | A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Technical Science |
| Publisher | Asian Institute of Technology |
| Abstract | Crime is an inevitable societal problem and is a major problem of any nation. Effective crime control requires the collection, organization and retrieval of a variety of data. Multiple types of data such as text (criminal data, property data, gang information, case information), graphic (photographs of criminals, pictures of crime scenes) and geographic (crime locations, details of the area) need to be accessed when and where they are needed, especially in time-critical situations. Therefore, there is a need to develop a system that provides accurate prediction for decision making in crime-control planning. This research was undertaken with the prime objective of developing a system for crime control. There are two main parts to the system: Internet GIS and crime risk factor analysis. In the first part, the system was developed using the Internet GIS technologies to prove its application in a flexible crime control environment. The system consists of mobile devices capable of accessing the Internet, with proper extensions for a GPS receiver and compact flash camera. Data can then be simply logged from the crime scenes and uploaded to a geo-database server. Open GIS enables spatial data sharing and system interoperability, which leads to data integrity and timeliness and reduces data replication. Open Source software and freeware packages, such as Minnesota MapServer, PHP, PostgreSQL and PostGIS, were used to develop the system. In the second part, the system for crime risk factor analysis was based on the crime pattern analysis of Brantingham and Brantingham (1991) and theory of crime control through environmental design. Processes were constructed to systematically analyse the factors affecting crime risk. The processes included the following five steps: (1) identify crime pattern characteristics, (2) establish the relationships between various crime factors, (3) determine the crime risk level, (4) recognize the crime data pattern using structured learning, and (5) predict crime risk factors. The model was constructed and tested using a Bayesian Network and Hugin software, which was also used to analyse the relationships within the data. From 150 samples, the data were divided into two groups: 70% of all data used for learning and 30% of all data used for testing. An empirical study on the predictive accuracy performance of the model is included. The Receiver Operating Characteristic (ROC) analysis was used to test the model and the results show the model performed an accuracy of prediction of 0.77. The developed system was demonstrated and tested with two sample groups, ninety Thai police officers and fifty Thai citizens. A questionnaire survey on the user acceptance of the system was conducted. The survey was designed to measure the acceptance of the system in terms of its usability to support crime control. The response received was positive from both police and citizens. Contributions to e-Govemment and e-Policing are (1) the study presented the system which is the first general framework for crime control and citizen services, (2) the study applied a Bayesian Network for crime risk factors analysis, for which no research work had been done before, (3) police can use the system to better control crime and provide public safety services for example, in Thailand, the system is beneficial to crime control policy makers and the operation of police in real time crime control, and (4) the model can be generalized to other public safety services applications in developing countries that have similar environment. For the future work, new Bayesian techniques will be explored to build a more complex and comprehensive crime risk analysis model which will include analysis of all types of crime |
| Year | 2006 |
| Type | Dissertation |
| School | School of Engineering and Technology (SET) |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Information Management (IM) |
| Chairperson(s) | Vatcharapom Esichaikul;Tripathi, Nitin Kumar. |
| Examination Committee(s) | Haddawy, Peter.;Masumoto, Shinji. |
| Scholarship Donor(s) | Ministry of University Affairs Royal Thai Government Fellowship |
| Degree | Thesis (Ph.D.) - Asian Institute of Technology, 2006 |