An Informative and Predictive Analysis of the San Francisco Police Department Crime Data
It is the responsibility of the San Francisco Police Department to protect the local community from various crimes and to improve the local security environment. With the development of modern statistics tools, we can learn from the past data and give suggestions for future strategy.
In this thesis, we study the San Francisco Police Department crime dataset from 01/01/2013 through 05/13/2015. Informative analysis regarding timing and location for different crimes are examined. Visualization methods are proposed for related features. We also discuss possibilities of predicting the crime categories given time and location data using the k-nearest-neighbor model and the logistic regression model.