Arrest Alerts #1

 

CONFIDENTIAL

DigiMastered Works LLC

Office of Development

DMW107IB1002

Date 07/14/2021 11:26PM

THIS IS FINALIZED INTELLIGENCE

 
To: The People
From: Tanner Fry
Subject: Arrest Alerts

 

About

    Arrest Alerts is a program written in Python that aggregates arrest data from the Missouri State Highway Patrol website. A request to scrape this data was put in to the department in 2019 and granted as long as any names were left out. Some of the data that is collected is as follows:

  • Age
  • City/state
  • Arrest date
  • Arrest time
  • Arrest county
  • Troop

An example of Arrest Alert adding new data.

Example of Arrest Alert adding new data

    Data can be collected on a single/continuous scrape basis depending on the function that is applied. The continuous scrape can be modified on how often a request pushed to the site. Currently, it is scraped every 30 minutes and then stores the data for later analytics. Some of the analytics we do is as follows:

  • Arrests by filter (Age, city/state, arrest date, arrest time, arrest county, and troop)
  • Mean arrests per day
  • Minimum and maximum arrests for a given filter
  • Total arrests/total days since scraping

An example of Arrest Alert displaying analytics

Example of Arrest Alert displaying analytics

Next Steps

    Arrest Alerts still needs more improvement. For instance, plotting of data is a priority so as to find information at a faster rate within the data. This can be done utilizing Matplotlib, a plotting library for Python. Other libraries that could be used are ggplot2, Plotly, and seaborn.

    Another improvement is to scrape actual charges, which are published on the site, from the each arrest and then aggregate that data with previously collected data to show insights for a given age, city/state, date, time, county, or troop. Through this aggregation we can gather a greater understanding of what kinds of issues the police deal with on our interstates, how often it happens in specific areas, and what kinds of charges are used the most.

Github

CONFIDENTIAL

Approved for release: 07/14/2021 ID: IB1002

1