11-632: Data Science Capstone - Syllabus
Time & Location
Section A: MW, 9:30 am - 10:50 am, TEP 1403
Contents and Schedule
Student teams will work on their assigned Capstone projects under the supervision of the project advisor and, where applicable, in collaboration with other students/faculty.
Class sessions are noted on the scheduled dates below. Fall plan feedback sessions, bi-weekly standups and midterm check-ins are to be held in-person during class meeting times.
There are three evaluation milestones, a draft, and final report, and a final presentation. There are multiple review assignments: bi-weekly standup reviews, midterm inner-team peer review, final inner-team peer review, and final team peer review.
Date | Activity | Deliverable Due |
---|---|---|
Aug 28 | Welcome Back / Overview | |
Aug 30 | Fall Plan Lecture | |
Sep 4 | Labor Day (No Class) | |
Sep 6 | Class meeting - Come for any questions regarding Fall plan | Fall Plan due Sep 7 |
Sep 11 | Fall plan feedback in class | |
Sep 13 | Fall plan feedback in class | |
Sep 18 | Weekly stand-up in class | |
Sep 20 | Weekly stand-up in class | |
Sep 25 | Weekly stand-up in class | |
Sep 27 | Weekly stand-up in class | |
Oct 2 | Weekly stand-up in class | |
Oct 4 | Weekly stand-up in class | |
Oct 9 | Weekly stand-up in class | |
Oct 11 | Weekly stand-up in class | |
Oct 16 | Fall Break (No Class) | |
Oct 18 | ||
Oct 23 | Midterm Check-in with Prof. Nyberg in-class Schedule TBD |
Midterm Grades due to University Oct 23 |
Oct 25 | Midterm Inner-team Peer Review due Oct 27 | |
Oct 30 | Weekly stand-up in class | |
Nov 1 | Weekly stand-up in class | |
Nov 6 | Weekly stand-up in class | |
Nov 8 | Weekly stand-up in class | |
Nov 13 | Weekly stand-up in class | Draft Report Due Nov 14 |
Nov 15 | Weekly stand-up in class | |
Nov 20 | Thanksgiving Break (No Class) | |
Nov 22 | ||
Nov 27 | Weekly stand-up in class | Draft Report Feedback will be returned by Nov 27 |
Nov 39 | Weekly stand-up in class | |
Dec 4 - Dec 13 | MCDS Capstone Final Presentation (schedule TBD) | |
Dec 14 | Team Peer Review Due Final Report Due Final Inner-team Peer Review Due |
|
Dec 20 | Final Grades due to University |
Assessment
The grade will consist of an assessment of both the quality of the data science experiment, its results, the technical process over the course of the semester as well as your evaluation by your peers.
The course grade will be based on the following:
- Fall Plan (required, feedback provided): 0%
- Bi-weekly Standups : 5%
- Bi-weekly Check-ins : 5%
- Bi-weekly Standups Peer Review: 5%
- Midterm Check-in: 10%
- The draft report is expected to be content-complete, i.e., no missing results, no missing sections, no text with grammatical issues as well as understandable diagrams and proper treatment of related work: 10%
- Final report is where you work out issues that arise during your final presentation, which will be graded as if it were submitted to a peer-reviewed workshop. You will also receive reviews from 11-631 students: 25%
- Final presentation: 30%
- Team Peer Review: 5%
- Midterm Inner-team Peer Review: 2.5%
- Final Inner-team Peer Review: 2.5%
Students are expected to take ownership of the project, take the initiative in driving the development forward and autonomously seek help when getting stuck. If requirements are unclear at any point, please talk to your project mentor or the instructors. For a detailed rubric of how the system, experiment, and results are assessed, you will be directed to the grading criteria document as posted on Canvas during the semester.
Technical Process Criteria
The Capstone project is also an exercise in proper software engineering. Your technical process evaluation will consider the following factors:
- Every project is required to use a GitHub or bitbucket repository.
- Every team member is expected to produce regular and sensible commits.
- To do items and nontrivial ongoing tasks are to be organized and documented in the GitHub issue system. This documentation is particularly important for planning milestones and action items produced during weekly meetings.
- Documentation must include a plan with timelines and milestones. Time and labor estimates for tasks are also a critical part of a project plan.
- Any documentation that will be needed on an ongoing basis (e.g., APIs, file formats, etc.) is to be kept in the GitHub repository readme and/or wiki pages.
- Code quality will be accounted for by the mentor and/or peer review.
Bi-weekly Standups
A standup is a casual check-in to share your team’s progress update. It is not supposed to be a polished presentation. Simply share your screen and brief us on what has been happening. Be authentic with what your team’s current progress is and what your team is dealing with. Doing so allows us to provide you with timely support if needed.
Each team is required to conduct a standup on a bi-weekly basis during class time. Students receive real-time feedback from the instructor and TAs.
Bi-weekly Standups
A standup is a casual check-in to share your team’s progress update. It is not supposed to be a polished presentation. Simply share your screen and brief us on what has been happening. Be authentic with what your team’s current progress is and what your team is dealing with. Doing so allows us to provide you with timely support if needed.
Each team is required to conduct a standup on a bi-weekly basis during class time. Students receive real-time feedback from the instructor and TAs.
Bi-weekly TA Check-ins
The bi-weekly check-in with the TA is a structured yet informal session to discuss your team’s progress, roadblocks, and future plans. This is an opportunity for a more detailed conversation compared to the standup, allowing your team to receive targeted advice and insights from the TA.
These check-ins are scheduled bi-weekly, outside of regular class time, at a mutually convenient time for both the team and the TA. You are encouraged to prepare a summary of what you’ve accomplished, where you’re stuck, and your next steps, but a polished presentation is unnecessary.
The check-ins aim to facilitate closer guidance, monitor team dynamics, and provide specialized input that may not be covered during the class standups. It also allows the TA to better understand your team’s progress and challenges, enabling them to offer more tailored assistance.
Students can expect constructive feedback and practical suggestions from the TA during these meetings, aiming to foster an environment of continual improvement and learning.
Peer Review of Bi-weekly Standups
The peer review of bi-weekly standups is an exercise in collaborative learning and constructive feedback. This process is designed to help you engage critically with the progress and challenges of another team while gaining an external perspective on your own team’s approach.
Each student will be randomly assigned to review another team’s bi-weekly standup, ensuring an unbiased and equal distribution among all projects. You will be informed of your assigned team via Piazza.
You are expected to watch the standup of the team you’re reviewing and prepare brief but insightful feedback on Persuall. The grading rubric for the standups is available to you. You can use it as a guide to provide feedback. Giving a score to the report is optional; however, the instructor and TAs will give the final score. The idea is to offer suggestions or observations to help the team you’re reviewing improve their project or approach to teamwork.
The deadline to submit your review is within 48 hours of the standup to give the reviewed team ample time to incorporate feedback before their next tasks or meetings.
Midterm Check-in
Midterm check-in is a standup-style, in-person meeting in class with professor Nyberg.
Inner-team Peer Review Assessment
To facilitate a fair distribution of work among team members, each midterm and final assessment includes a peer review. It includes
- Evaluation of one’s own and other’s contributions to the project.
- Comments about my own contribution to peers.
- Comments to peers about peer contribution.
- Comments to instructors about their own and others’ contributions.
Team Peer Review
Each student is required to attend at least one final presentation of another team and submit a brief review of this team’s presentation. Students will have the chance to submit their preference for the team to review according to their schedule suitability.
Academic Integrity
For all presentations and the final report, students share work with their teammates. Both presentations and reports need to make clear at all times the students’ contributions and those parts that have been influenced, taken, adapted, or otherwise derived from prior work. Any such additional material must be properly cited in the report and on the presentation slides. This citation material includes related academic writing (even if published by a collaborator of the project), diagrams, datasets, prior Capstone project reports, video tutorials, scientific blog posts, and technical components such as algorithms, libraries, and the like. When a source’s text is paraphrased, it needs to be referenced. If it is reused verbatim, it must be quoted.