Crime is expanding all around the world through many ways, and violent crimes have become a national dilemma. Family members try to ignore, and neighbours do not want to get involved. The victim in such case is usually helpless to understand and to determine the solution of the problem. Also, law enforcement agencies handle each case when someone got affected or when a criminal act has taken place. Data mining turns an essential role in the prediction and the analysis of data. For the crime analysis in various states of the United States, several classification algorithms have been applied to the crime data of the FBI concerning the population of states. Experimental result showed that two classifier algorithms Reduced Error Pruning Tree (REP Tree), and Naïve Bayes produced better results as compared to other algorithms. However, most of the results of Classifier algorithms were same, so we concluded that our dataset had been classified correctly. The outcome from the analysis shows that crime rate is high in populated states. It is observed that rate of property crime and rate of larceny-theft is high amongst all other types of crime rate in populated states of the United States of America (USA).