The identification of causes and phenomena associated with crime is one of the mo s t popularrngoals in criminology, especially in view of its practical value and the belief that suchrnidentifications are useful w hen seek in g to correct or control criminal behavior.rnThe utility of discovering causes must, however, be qualified. Understanding and processing ofrnoffenders' record s is one method to learn about both crime and the individuals who in valve inrnmisdeeds so that police can take crime prevention measures accordingly.rnThough data on criminals are continuously being gathered, the y are not effectively being utilizedrnfor extract ting pattern s that can be used for effective management of crimes. This is mainly due tornthe inadequacy of the human brain to search for complex and multifactor depend envies in datarnan d the lack of objectiveness in such analysis demanded a computerized approach.rnDevelopments in the inform at ion and communication tech neologies have made it possible forrnorganizations to collect, s tore and maniple ate massive amount of data. One such development isrnBayesian Network.rnIn this study, the main objective of the research is to develop a predictive model for factors thatrnconstitute higher crime trend s in Addis Ababa which makes use of Bayesian Network modelingrntechniques. For this purpose, published literature's in related areas have been studied togetherrnwith the review of different Bayesian Network modeling app roaches . Different to oils andrntechniques supporting such task were examined by taking into co n side ration the reapplication tornthe problem domain. In addition, an experiment is conducted to explore the potential of Bayesianrnnetwork in modeling factors that constitute higher came trend using personal identificationrnrecord of criminals.rnFor the purpose of the experimentation o n 1572 criminal reco rd s were collected from the AddisrnAbaba Police Commission. The record s were manually and automatically y further p reprocessed tornmake them compatible with software used. Important attributes that are considered relevant forrnthe construct in g predictive model for higher crime trends were selected.rnAfter p reprocessing the data, alearnin g classifier is used to learn from the training data and usernthis classifier to class if yew data. A model is constructed for the best learned model from data.rnBased on the experimental data, a Bayesian performance prediction model was developed wherern73.25 % prediction accuracy was fir stobse rved . Further experiments and modification of thernprediction model in creased the level of prediction accuracy to 75.78 %.rnFin all y, Three Phase Dependency Analysis in particular and Bayesian network in general isrnfound applicable for modeling determinant factors for higher crime trends.