Law enforcement agencies like that of police today are faced with large volume of data that mustrnbe processed and trans faced into useful intonation and hence data mining can greatly improverncrime analysis and aid in reducing an d preventing crime.rnThe purpose of this study is to explore the applicability of data mining technique in the efforts ofrncrime prevention with particular emphasis to the Oromia Police Commission n an d to build arnmodel that could help to extract crime patterns. With this objective decisis on trees and neuralrnnetwork were employed to classify crime records on the basis of the values of attributes crimernlabel (Crime Label) and crime scene (Selle Label). Results of the experiments have shown that decision tree ha s classified crime records at anrnaccuracy rate of 94 percent when the attribute Crime Label is used as a basis for class fiction.rnWhere as, in the same experiment, the accuracy rate of neural networks is 92.5 percent. On thernother hand , in the case of classification of records on the values of the attribute Scale Labelrndecision tree h as shown an accuracy rat e of 85 percent while neural network revealed 80 percent. In both experiments the output indicated that decision tree performed better. Besides, decisionrntree generated understandable rules that could be easily presented in human language and thusrnpolice officers can make use of these rules for designing crime prevention strategies. Thus, thisrnexperiment has proved that data mining is valuable to Support the crime prevention process andrnparticularly, decision trees seem more appropriate for the domain problem.