CS 638: Data Science in the City of Madison
If you want to enroll in this 2-credit project based course, (a) join the waitlist online, making sure you enroll in the section taught by Tyler Caraza-Harter, and (b) and fill this form. I'll let you know if you can take the course by Aug 28th. We might not have room for everybody interested, unfortunately. The prereq is CS 301. If you haven't taken that, but have experience with Python (including Jupyter, pandas, and matplotlib), I may override the prereq; in this case, please email me (firstname.lastname@example.org) a Jupyter notebook file demoing your skills. You'll need a laptop that you'll use to present your work to the class on a regular basis (you might need to buy a video adapter for the room if you don't have an HDMI port).
Time: 4-6pm on Thursdays.
This course will involve writing Python code to analyze and visualize data sets about the City of Madison, WI. The goal will be to arrive at actionable suggestions. In 2012, the city adopted an open-data policy; projects will analyze data that is released under this policy (http://data-cityofmadison.opendata.arcgis.com/) and possibly other data that city collaborators may provide for this course. Project teams will create data visualizations and present them to the class on a weekly basis. Teams will choose their projects and set their own research agenda. Possible topics include, but are not limited to, (1) exploring alterations to metro routes to improve efficiency, (2) identifying ways to reduce traffic accidents, (3) projecting property tax revenue, (4) simulating changes in salary spending across city departments, and (5) identifying ways to reduce incident response times for the fire and police departments.
Students will form teams of 2-4 people and choose a topic. Within each team, each of you will be expected to produce 2-3 high-quality visualizations per week about your chosen topic; the code and plots will be committed to a team git repo. You will state the question(s) that motivate each plot and explicitly write their conclusions; you will further organize the plots into a logical narrative sequence. Each week, you should be prepared to present your weekly analysis to the class; depending on enrollment, you might present every week, or I may randomly select a subset of the class to present each week. At the end of the semester, each team will consolidate their results into a full presentation. You will be graded on (1) your visualizations, (2) your communication about those visualizations, (3) your participation in discussing the results of other students, and (4) your final presentation.