Welcome to CS184
Bridging Policy and Tech Through Design
Thanks to Coronavirus ...
- Proposal deadline deferred to Sunday, March 29.
But please get those proposals in as soon as you feasibly can.
- And please tell us in which time zone you expect to be (so we can make sure an on-line team meeting is realistic).
- Class lectures will still be held at the regular class time, Tuesdays, 4:30PM, but they will be
recorded and there will be on-line chats to supplement discussions at the lectures.
A Project Course
This course is intended for teams of 3-4 students, with at least one from the social
sciences and at least one interested in computing.
We have received 28 applications and everyone has been placed on a team. Most people got their first choice, but we did
have to give some a second choice in order to balance out the teams.
There are a number of constraints, including the fact that we need to have people with
different majors on each team, and we can't have a team so widely dispersed around the globe that it is impossible for them
to find a time for an on-line meeting when all are awake.
You will receive an email introducing you to the rest of your team and at least one mentor by the end of Wednesday April 1
Because the quarter has been cut short, it is important that each team start to discuss the details of their project plans
before the first class meeting on April 7.
If there are any questions, please contact the instructor (ullman at gmaildotcom) or any of the course staff.
- The course will meet once a week, on Tuesdays at 4:30PM for at most 80 minutes.
- The room for these meetings is 300-303.
- Each team will also meet at least once a week, at a time of mutual convenience, with
at least one mentor.
- NB: we are prepared to deal with coronavirus by making all of these meetings virtual, if
Teams are encouraged to solicit the involvement of a faculty member
who is knowledgeable in the domain relevant to the project, but the course staff will
attempt to help with that process and will also make certain there is a mentor for the
data-science aspects of the project.
There will be a midterm progress report from each team at the 5th class meeting, and
a final oral report at the final meeting.
Teams will also be required to file a written report on the conclusions drawn from
We are in contact with a number of partners who are willing to share data.
Links will be added to this list as soon as we are able.
You can search all ads at Facebook
Here's an example of what can be known about a particular ad:
In addition, there is a repository of TV political ads that
might be of use in a project about ads.
- Southern Poverty Law Center
This is a collection of far-right chat servers, including the servers that were used to organize the 2017 white supremacist
rally in Charlottesville: Discord Leaks.
This is an influential forum for sharing extreme right wing ideology:
Iron March Exposed
This is a database of articles and other materials about human-rights violations:
California Legistative Actions.
Here are a number of datasets for various projects proposed by Stephane Dugain:
This is a Sample of 5K from a set of 60K
survey responses from those served by nonprofits.
- Human Rights Data Analysis Group
They have a number of collected datasets that can be accessed Here.
Note: this group is not proposing a project, but their data could be interesting to people working with other partners or a
team that wishes to create their own project.
- Trump Twitter Archive
SQL is the standard language for querying databases. It is much more powerful than spreadsheets, as well as being able to
handle really huge files. Jennifer Widom, the Dean of Engineering, created a MOOC that covers the elements of SQL
programming. Alas, it is in transition between Stanford's own MOOC platform and EdX. I hope it will be available in time
for people who need it to take advantage, but for the moment, she suggests people try one of these tutorials:
If you need to run a SQL database system, Microsoft Access is a possibility. There is also a free SQL engine called MySQL. Another option is Postgres. Any of these will be suitable even for gigabytes of data, running on
your laptop. If anyone has a need to process really large amounts of data, we can probably :arrange for free Google cloud service.
Email the instructor, Jeff Ullman (ullman at gmail dot com).