This is Ziwen Lu, Xuyan Xiu, and Doris Yan's machine learning project to predict 2016 crime rate using 2010 as the modeling year, and we compare the actual 2016 crime data with our predictions. Please click on GitHub Pages for details.
Our project contains four parts:
- full write-up, including introduction, model and methods, results and limitations of our project;
- geospatial mapping of average crime rate from 2002 to 2014;
- predictive modeling of 2010 crime rate on 2016 crime rate;
- slides for presentation.
Our models are robust to predict the unseen data. Applying our model of violent crime rate on the implementation data in 2016 generates RMSE of 0.593, which means the average difference between our model's predicted values and the actual values is 0.593. The unit for RMSE is log crime rate per 100,000 people. Similarly, for property crime rate, the average difference between our model's predicted values and the actual values is 0.459.