Crime Forecasting By Using Data Mining Techniques: The Case of Addis Ababa Police Commission
| dc.contributor.advisor | Kebede, Gashaw (PhD) | |
| dc.contributor.author | Negussie, Hailemariam | |
| dc.date.accessioned | 2018-11-28T08:04:57Z | |
| dc.date.accessioned | 2023-11-29T04:56:58Z | |
| dc.date.available | 2018-11-28T08:04:57Z | |
| dc.date.available | 2023-11-29T04:56:58Z | |
| dc.date.issued | 2015-10 | |
| dc.description.abstract | Law enforcement agencies like that of police today are faced with large volume of data that must be processed and transformed into useful information and hence data mining can greatly improve crime analysis and aid in reducing and preventing crime. Knowledge discovery process has come across a variety of approaches. Hybrid methodology of Knowledge discovery process is one of the popular approaches used in Knowledge discovery process models. Hybrid knowledge discovery process model is a problem solving strategy for this study. The purpose of this study is to identify the relation between offenders, victims and offences and to develop crime prediction model from the available data of Addis Ababa police commission. With this objective, J48 decision trees and PART rule induction algorithms were employed to generate patterns and classify crime records on the basis of the values of attributes CrimeLevel, Time, OffenderJob, Victimjob, OffenderMartialStatus and VictimMaritalStatus. Results of the experiments have shown that PART rule induction algorithm has classified crime records at an average accuracy rate of 88.6% when the above stated attributes were used as a basis for classification. In the experiments, the output indicated that PART rule induction algorithm performed better. The model has been evaluated on the testing dataset for crime level target class and scores a prediction accuracy of 97.2%. Besides, PART generated understandable rules that could be easily presented in human language. The research also demonstrates that crime type, age, educational level, job, marital status, sex, and particular place were the common demography factors of victims and offenders who were exposed to crime. The challenge in this study was, since the demographic similarity of both victims and offenders. So, further study has to be done for each alone. | en_US |
| dc.identifier.uri | http://etd.aau.edu.et/handle/123456789/14596 | |
| dc.language.iso | en | en_US |
| dc.publisher | Addis Ababa University | en_US |
| dc.subject | Data Mining Techniques | en_US |
| dc.title | Crime Forecasting By Using Data Mining Techniques: The Case of Addis Ababa Police Commission | en_US |
| dc.type | Thesis | en_US |