CS 838 Spring 2019 (Also Known as CS 839)

Data Science: Principles, Algorithms, and Applications

WedFri 2:30-3:45pm in 168 NOLAND, 3 Credits

Announcements

• The class mailing list compsci839-2-s19@lists.wisc.edu. It is very important that you are on this mailing list. You should be there automatically if you are enrolled in the class. But do keep an eye out for whether you receive emails from the list.

  • The class's Piazza page is available here.This is a forum for the students. We will monitor occasionally but do not have enough man power to answer all questions posted to this page.

Instructor & TAs

• AnHai Doan, contact information available from my homepage.

• Office hours: Wed 5-6pm and by appointment (pls send email, thanks)

• TA: ARAVIND SOUNDARARAJAN <soundararaj2@wisc.edu> , office hours: to be decided.

Course Description, Prerequisites, and FAQs

Course Format & Grading

• • See the above course description

  • Midterm: March 29 Fri, in class at usual time/room

  • Final: May 3 Fri, in class at usual time/room

  • Grading: midterm: 30%, final: 30%, project: 40%

Lecture Slides (tentative)

Note: the slides below are those from the previous offering of the course. I will update slides AFTER the lectures. When a slide set has been updated, I will indicate so.

• • Introduction (updated)

• Background: tools, data models/formats, and attributes

• Problem definition and data acquisition

  • Data extraction

  • Information extraction from text

    • Extraction from template-based data (aka wrapper-based extraction)

• Data exploration, profiling, cleaning, transformation

  • Data integration

  • String matching

  • Data matching

  • Schema matching and putting it all together

  • Data exploration and analysis

  • OLAP

    • classification, clustering

    • association rule mining (see the book chapter)

    • anomaly detection

• Building data-intensive artifacts & designing data-intensive experiments

• Cross-cutting issues

  • • cross-cutting techniques, execution stages, workflow management, team organization

    • the three Ss: stages, steps, stacks

    • scaling, quality monitoring, crowdsourcing, etc.

    • implementation/architectures

Project

• • Students will form teams for a multi-stage project that addresses a data science problem.

• Project page

Resources

• • Click here for resources to learn Python, pandas, machine learning, more data science, etc.

• If you want to study more data science at UW

Misc

• • dotdatascience.org UW-madison student organization focused on data science