This is the webpage for CMSE495 Data Science Capstone Course. These materials are provided as an Open Educational Resource (OER). Instructors interested in using these classroom resources should reach out to Dirk Colbry (colbrydi@msu.edu) who can provide all the materials and instructor notes.
Replace the following with a picture that “defines” your project. This could be sponsor logo, an expected outcome of your project, a graphical representation of the research area, etc.
Image from: URL REFERENCE
Please write your proposals as word documents. This page not only acts as instructions for your proposal but should be used as a general outline for writing.
Write the proposal to a general audience that understands data science but doesn’t know about the course or the project (other data science students, your other professors or your roommates).
Make this look as professional as possible. You will be sending this to your project sponsors.
Include a very brief description of the course and the course goals (similar to what you should have put in your Git repository README file). Then provide a couple of paragraphs about the problem you are going to cover in your project. Make sure to include research questions and goals. Explain as if to someone outside your area of expertise. Please use references when appropriate.
In this section include a technical description of your planned solution. What format is the data in? How will you read in the data? What language will be used? What algorithms and programming libraries will be used? How will the output be formatted? How will you generally measure success (more details about success measures can be made below)? What computers will be used to run the code? How will the code/results be shared with the sponsors?
Many of these will be open loops. You can describe items in this section as an “open loop” and then flush out the details about how you will close the loop in the project goals section below.
Explain in more detail the major challenges and components your team needs to get working for the project success. Specifically focus on “Open Loops” you need to close and “Metrics” you are going to use to measure success.
Use the term “open loops” to describe problems for which you know there is a solution (or can find the solution) but you will need to go through some sort of selection progress to figure out how you will move forward. For example, your project may require a “machine learning model” but your group will need to review the various models and pick one that best fits your problem. Selecting the model you will use in your final project would be “closing” this open loop.
All projects have open loops until there is a working end-to-end prototype. You want to get to your end-to-end prototype as soon as possible and then iterate on the design. Include in this section the main topics you need to figure out to get to your end-to-end solution. List them here and provide a plan or selection criteria you are going to use to close the loops.
Include a section about how your team will measure success. How do you know one model is better than another? How do you know if you are done. Whenever possible try to pick something that is quantifiable (time, precision, etc). Without a measurement you will not be able to evaluate progress toward your goals.
Now list your team members, their strengths and how they are planning to contribute to the project. Include individual goals each member of the team would like to get out of the project. For example, skills they would like to practice or learn.
How will the team communicate with each other? At what interval will the team communicate with the sponsors (weekly meeting, weekly email)? What are the backup plans when communication goes down?
Build week-by-week timeline of the project (include 1-2 items for each week of the semester). You can use the course milestones as a guild but try to focus in on project specific goals such as when specific open loops are going to be closed and when prototypes for this project will be completed.
You instructors recommend starting at the end of the semester and working your way backward through your goals. This will ensure you don’t pile everything up at the end of the semester.
List any key pieces of information/skills do you need to learn and/or master to successfully complete all of your project goals.
Also list challenges that you are worried about or that may come up in the course of completing your project. Include changes you will make to your plan if these challenges do arise. Try to anticipate things that will happen (such as illness) and account for these challenges. Make “backup” plans that can catch many challenges that you can’t anticipate.
The following basic grading rubric instructors will try to use to evaluate your proposal
Grading Overall
10 points - Project title and Picture
20 points - Technical Details
20 points - Overview
20 points - Project Description
20 points - Division of Labor
20 points - Anticipating Challenges
Grading Rubric
-5 Leaving in instructions in report.
-5 Any Sloppy formatting
Add the proposal to your Team folder with the completion date in the title. For example YYYYMMDD-Proposal.docx. The file location should be obvious in your folder structure and the instructor should not have to hunt it down.
Once your instructor has reviewed the assignment you will be expected to share it with your sponsors.
Written by Dr. Dirk Colbry, Michigan State University
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.