kjs222 / police_data Goto Github PK

• make City of San Diego police incident data accessible, searchable and useable via API as well as access to a relational database dump file.
• help members of the public build effective search queries through interactive API documentation
• help users visualize the data with charts

This project represents my final project at Turing. The project spec required that we pursue new technologies that were outside the curriculum. I chose to develop an app completely in Node/Express - with both server side and client side Javascript. I also chose to explore D3 and libraries built on top of D3 for data visualization. This was my first time using all of the following technologies:

• Node/express
• Sequelize as the ORM
• D3 for charting
• Dimple JS for charting
• Node Cache Manager to cache data for charts
• Swagger for API documentation

Note: D3-only is a branch with some WIP D3-only charts (i.e. no Dimple). I plan to keep working on these and see if I can recreate all of the charts only using D3.

Production Site | API documentation

API endpoints:

/api/v1/incidents
/api/v1/beats
/api/v1/dispositions
/api/v1/call_types  

the /api/v1/incidents endpoint takes a number of optional search params:

number
priority
beat
neighborhood
disp_code
disp_desc
call_code
call_desc
street
street_number
date
start_date
end_date

Resources: raw data sets - CSV format | PG database dump - TBP can import into a postgres DB

The CSV data is updated weekly from the City of San Diego. The data required extensive cleaning and normalizing for the purposes of this project. I have notes on the cleaning and normalizing process that I followed, which I would like to include on the project site. I would also like to identify a process for routine updates to the database based on updated CSV data from the City.