Navigating Big Data

This activity embeds research, problem solving, and presentation skills to expand prior knowledge on big data. 

Goal: To teach students how to brainstorm, how to creatively think about research and how to communicate fundamental computer science concepts via discussion.

Background:

In this activity, the instructor chooses the papers to be discussed and the concepts that are important to highlight. The readings can be found on the class website: http://daslab.seas.harvard.edu/classes/cs265/. The students then get access to the papers and read 1-2 research papers prior to class and are tasked with writing a review of the papers.

Procedure:

  • In each class a discussion is initiated via a presentation of 1 or 2 students of the papers they have read.
  • As the presentation progresses the discussion develops around the topics presented. In every slide presented the instructor will initiate the discussion if students do not do so themselves.
  • The students propose ideas to problems that the instructor has highlighted and bring up new problems for discussion with the whole class.
  • This activity is essentially a big brainstorming meeting driven by the topic of the day with 10 specific questions the class is trying to answer:
    • what is the problem of the paper discussed?
    • why is it important?
    • why is it hard?
    • why existing solutions do not work?
    • what is the core intuition for the solution?
    • solution step by step
    • does the paper prove its claims?
    • exact setup of analysis/experiments
    • are there any gaps in the logic/proof?
    • possible next steps
  • Next, the instructor summarizes concepts learned and the main points to remember.
  • The students also use the lessons learned to inform the design of their semester research project.

Comments:

  • The instructor advises that communicating complex concepts via discussion is much more approachable to students than just lecturing. According to the instructor, discussion gives immediate feedback to the professor about the level of understanding of each student and the kind of feedback they need.
  • The instructor also advises to carefully pick the topic of research papers and carefully prepare a list of discussion triggers for each class to make students think of the right concepts for discussion

Assessment: The students produce a software deliverable that is designed using principles learned in class. They deliver a demo and a presentation of the final result.

Submitted by Stratos Idreos, Computer Science