One potential approach for crime analysis that has shown promising results is data analytics, particularly descriptive and predictive techniques. Data analytics can explore former criminal incidents seeking hidden correlations and patterns, which potentially could be used in crime prevention and resource management. The purpose of this research is to build a crime analysis model using supervised techniques to predict the arrest status of serious crimes in Chicago. This is based on specific indicators, such as timeframe, location in terms of district, community, and beat, and crime type among others. We used time series and clustering techniques to help us identify influential features. Supervised machine learning algorithms then modelled the subset of features against incidents related to battery and assaults in specific timeframes and locations to predict the arrest status response variable. The models derived from Naïve Bayes, Decision Tree, and Support Vector Machine (SVM) algorithms reveal a high predictive accuracy rate at certain times in some communities within Chicago.