A look into the primary crime type of WEAPONS VIOLATION from Chicago’s crime dataset on data.cityofchicago.org. Primary Type of weapons violation includes such descriptions as; DEFACE IDENT MARKS OF FIREARM, POSS FIREARM/AMMO:NO FOID CARD, RECKLESS FIREARM DISCHARGE, UNLAWFUL POSS OF HANDGUN, and others.
Multifamily public housing development visualizations by unit count/neighborhood for multifamily property types, and a word cloud of the most popular management companies. The data is from data.cityofchicago.org
The affordable rental housing developments listed below are supported by the City of Chicago to maintain affordability standards.
The data was found on https://data.cityofchicago.org/. The primary type was filtered by homicide. Date range from January 2001 to present was used.
The data is from data.cityofchicago.org, and represents a time range from 2014/7 through 2015/9.
655,074 total violations.
A different take on Chicago Tribune’s Gun Shooting visualizations. Time Range: JAN. 1, 2015 through AUG. 31, 2015.
The data was obtained and parsed from the Trib http://crime.chicagotribune.com/chicago/shootings
I made a daily line chart using Google Charts, and an overall intensity map using CartoDB.
I found an interesting dataset on political contributions in the state of Illinois.. The downloadable .zip contains multiple tab delimited database files which contain the relationships between Donations, Committees, and Candidates.
Out of curiosity in seeing Comcast’s political influence in Illinois over time, I parsed the 650,775mb file called Receipts.txt. Below is a bar chart of yearly recorded donation totals from 2000 through 2015-08.
Not being very politically oriented, I wanted to somehow relate the donations to candidates. But in the form of the available data, it appears Donors make contributions to Committees, and Committees support a Candidate. But I do not know if Candidates and Committees are a One to One relationship at the time of typing this.
Parsing the text file called CmteCandidateLinks.txt, I related the Committee Id with the candidate Id. Parsing the text file called Candidates.txt I relate the Candidate Id to the Candidate name.
Lots of candidate duplicates per donation entry. Majority of Committees represent the same candidate under different ids, while some committees represent multiple candidates. Example here:
So I decided to distribute each donation amount between a potential multitude of candidates. I did this by dividing each donation by the number of candidates which belong to the committee recipient. From that, I got this list of Comcast’s Top Illinois Candidates.
I came across an interesting data set while browsing datacatalog.cookcountyil.gov. It can be found here. Judging by the title, it appears to be healthcare costs for the month of June in the year of 2012. There is no description, or any additional information I could find about it. The data contains a department code, a number of employees, and a total cost. It looks like this:
I was curious, so I took the average cost of employee count (Cost / Emp num), per department. I correlated the department numbers to labels from an alternative dataset also found on cook county’s data catalog.
My experiment resulted in two bubble charts.
Greater than 40 Employees
Greater than 40 Employees and less than 1,000
And now here’s the second chart processed with Google’s Deep Dream
A look into Chicago Police Department’s incident reports for marijuana related arrests.
The chart below displays over 15 years of CPD’s Cannabis related arrests for possession of 30 grams or less. The data is from data.cityofchicago.org
Here is the ruby code written to accumulate the monthly totals. Additional data filtering was done on the Socrata website prior to exporting the csv data file. Google Charts was used to create the line chart. CartoDB was used to create the still of the torque map.