Using Machine Learning for Prediction of Factors Affecting Crimes in Saudi Arabia

Crime rates are expected to increase in the whole world as the growth of many complex factors like: unemployment, poverty, weather, violent ideologies and religion and etc. Obviously crimes have negatively influenced the development of society, economic progress and reputation of a nation. Hence, Analyzing large volume of data with machine learning algorithms can be used to predict the crime distribution over an area to provide indicators of specific areas which may become a criminal hotspot. The aim of this paper is to predict factors that most affected crimes in Saudi Arabia by developing a machine learning model to predict an acceptable output value. Our results show that Factor Analysis of Mixed Data (FAMD) as features selection methods showed more accurate on machine learning classifiers rather than Principal Component Analysis (PCA) method. Naive Bayes classifier perform better than other classifiers on both features selections methods with accuracy 97.53% for FAMD and PCA equals to 97.10%.