Analysis of San Francisco 311 and Police Department data
Kaggle hosts a lot of great publicly available datasets, and one of them is San Francisco Open Data (BigQuery Dataset) which was available since 2018. I've seen quite a few notebooks utilizing the data from this dataset, however most of them were focused on bike sharing data.
My interest however was in the public safety area, so I wanted to analyze information in the 311 reports database and the SFPD database of this dataset. The first goal of this project was to get a sense of what is the most pressing issues the citizen of SF were concerned about, to find correlated issues, and to produce an interactive map of the city reflecting geographical distribution of the issues.
The second goal was to analyze SFPD data and determine longitudinal and geographical dynamics of the crimes.
And lastly I wanted to visualyze the relations between the information from 311 reports with the crime data.
The information in this dataset is stored in BigQuery format, so naturally I used BigQuery API to access the data via SQL. And to visualyze the results of the analysis I used Folium to build Choropleth maps and heatmaps.