[EPIC] Data Science Insight Discovery using 311 Data (Research) H1 2023

[nichhk]

This is the Epic for all Data Science Research Projects

All data science team members may feel free to add your projects to the ongoing task list, below! If you have a project repo or public link you'd like to use to share your project work, make sure that link is referenced in your project's issue.

Overview

To generate impact, I (Nich) think there are a two main ways of thinking about these Data Science Deep Dives (DSDDs):

• investigation/justice-oriented: is the city treating and serving Los Angelenos with equal treatment?
• Efficiency-oriented: how can we make city services and processes more efficient?

(If anyone has other ideas on how to frame our DSDDs, please comment below.)

We've already created other issues to track specific DSDDs, with 1-2 data scientists working on a particular DSDD. So this issue just serves to track our overall progress and work towards the main goal of producing at least one interesting, actionable insight--the more the better!

Action Items

Investigation/Justice

• Investigate relationship between NC socioeconomic class and 311 requests #1278
• Investigate the relationship between crime and 311 requests #1277
• Investigate relationship between air quality data and 311 requests  #1294
• Research flood data/respond to UCI researcher #1098

Efficiency

• Identify addresses with significantly more 311 requests #1279
• Develop a 311 request forecasting model #1289
• Research incorporating population data into 311-data reports/dashboards #1229

Both

• 311Data: 311 request types with no council (councilId=0) assigned to it #1250
• Data Science: Create a forecasting model for agency resource insights #1432

Backlog

• Identify the regions to set up LA Metros Bike Share Stations to help mitigate "First/Last Mile Problem"  #1292
• Investigate the relationship between car parking meter utilization, number of trees, and 311 Requests #1293

Resources/Instructions

• Get your 311 datasets here with the API. A data range may be entered using the CLI. If you need help with this ask a Lead or a Peer! This may generate a large dataset so be prepared. For example, one year may have 1M+ data points (individual 311 requests).

• City of LA 311 data resources

• Shape files for Neighborhood Council here

• Census city block shape files here (tabblock20)

• If you want to do an analysis using Census blocks, use this script on the CLI to do a spatial join with Census block IDs and your API dataset from above.

• Here is a list of Data Sources to generate some ideas for projects.

• Staying informed about what's going on in LA is also helpful--knowing the pain points for Los Angelenos will help us to know which problems to solve.

[nichhk]

Tagging all our data scientists: @joshuayhwu @susanklm @piotrsan @priyakalyan @Michaeldavidng @JACKIEKWON @lgewirtz @yn2191

[ajmachado42]

Other research projects exist that weren't included in the backlog section of this issue. See Data Science role tag for those other projects.

Adding a note at the top of the description that communicates this is an ongoing Epic and that any team member may update with new research projects and upon completion of projects.

[ajmachado42]

Reopening since this is an ongoing Epic.