About this document
San Francisco Crime Prediction Analysis by José Eduardo Mundarain Gordillo is a document available to read on EtoBox.
This document discusses analyzing a dataset from the San Francisco Crime Classification competition on Kaggle. It finds that hour of day is the most important feature for predicting crime category. It also examines the distribution of crimes by police district, day of week, and whether they occurred at a street address or block. Principal component analysis was used to reduce the number of features and avoid overfitting, finding that using the top 3 components worked best.
- Author
- José Eduardo Mundarain Gordillo
- Language
- EN