This page is for information related to data science competitions.

Planning for first UW Data Science Center analysis competition  

In fall 2019, the Data Science Center will host a data science competition at UW.  Here are some thoughts about what we might do, to solicit feedback and input from participants in the center.  This is the working draft of ideas, as we gather input and learn about good features of competitions elsewhere.

• We would release a description of the competition, which would be open for several weeks, with a deadline for submissions.  We will likely pre-announce the subject area and type of analysis, so that competitors can prepare and learn about appropriate tools before the competition opens.
• The goal would be an analysis of some real data that pertain to some phenomenon that is relevant to Wyoming.  The analysis could be a prediction problem, or a more general analysis problem.
• Entries into the competition would be judged and ranked according to four criteria:
  1. Accuracy of the prediction
  2. Accuracy of the explanatory components of the analysis, in which importance of variables in the prediction is reported
  3. Repeatability of the analysis (or more generally process, including communication)
  4. Quality of visualizations and effectiveness of other elements of the presentation of the analysis
• We have financial support for cash prizes for winners of categories, as well as possibly a best overall analysis winner.
• We will seek knowledgable representatives from the business, state agency, and academic entities to serve as judges and discussants of the analyses at a presentation, judging, and prize event.  We will invite additional attendees to listen to presentations and interact with teams that enter analyses in the competition.
• The competition will be open to teams of students, postdoctoral researchers, and staff at the University of Wyoming and all community college in the State.  Teams may be assisted by university and college faculty and staff.  It seems likely that teams will spend 20-100 hours on entries to the competition.

Lightning round competitions

Lightning Round Competition Brainstorming page

Kaggle data science competitions

• Titanic: Machine Learning from Disaster
• Data science bowl 2018