Final Project
Overview
You will be working with a team of students (2-3) on a final project for data science. We will be doing some ideation on Friday, March 11th as well as more ideation /team formation in class on Tuesday, March 22nd. The project can be centered around or bridge the major themes of this course.
Project Topic Guidance
You have quite a bit of autonomy. Here are some general categories of projects:
Implement some data science algorithms. The aim would be to learn more about how these algorithms work, and possibly to create some sort of library / tool for others to use.
Expand on the themes of the CTW project (data visualization, infographics, and data journalism).
Work with an external or internal (meaning at Olin) collaborator. This would involve acting as a consultant to the collaborator: exploring some data, answering some questions, forming interesting hypotheses.
Do some more machine learning. This could include working on Kaggle competition(s), exploring some advanced algorithms (neural network), etc.
Develop a data-powered app or tool. The options here are limitless. Pick a user group, and then figure out how data might help them, and how an app might put this power in their hands.
Other random thoughts on projects:
I'm okay with projects spanning multiple classes (OlinJS comes to mind), as long as that is okay with the teaching team for the other class.
We will be doing a scaffolded activity that invites you to think about your learning goals for the final project, however, in advance of that make sure you are thinking about how your learning goals align with the types of projects you are thinking about.
Teaming Considerations
We'll be doing an in-class team formation activity on Tuesday, March 22nd. The only ground rule I have at this point is that the teams must be either 2 or 3 students.
Project Deliverables
Project Proposal (0%)
Poster / demo (10%)
Final output (encompasses functionality, design, documentation) (30%)
Website. Includes project stories (or blog) and project documentation. (10% for blog entry 1, 10% for blog entry 2, 10% for the final project documentation).
YOGA (30%)
YOGA
Phase 0
The day before the in-class project ideation and team formation activity, you will write up a list of three goals that you wish to achieve during the final project. Your goals can take any form as long as they represent your own interests and learning objectives for the final project. You are welcome to build on the course learning objectives, or you can strike out in a new direction. Make sure that each goal is written in a clear manner: the goal should be something that you can achieve, and you need to be able to know if/when/how well you have achieved it.
The phase 0 submission should be a list of three goals, each one about a paragraph (3-4 sentences). Explain what you hope to do or learn, how you will achieve it, and (briefly!) why it is important. To turn your assignment in, you will share a Google doc with me (paullundyruvolo@gmail.com). Please name your document "Lastname, Firstname DataScience YOGA" so that I can keep track. This is due 3/22.
Phase 1
Once you are on a specific team and have written your project proposal, you may want to revise your goals. This is your chance to change your goals for whatever reason, or perhaps to clarify them. If you are happy with the goals then you can leave them unchanged. Whether you revise them or not, at this time we want you to add an assessment plan for each goal. How will you (and we) know that you achieved each goal, and how will we know how well you achieved it? The assessment plan does not need to be a tangible assignment; you are welcome to propose a reflection or any other strategy. To turn in this assignment you should add a new page to the Google doc that you have shared with me. Please don't delete old content, but instead append phase 1 at the end (even if it is largely the same as phase 0). This is due 3/25.
Phase 2
About the midpoint of the project, you will have an opportunity to revise your goals and/or your assessment plan. Please append a "Phase 2" section to your Google doc that states your potentially revised goals and assessment plan. At the top of your phase 2 section, please briefly (maybe 2-3 sentences) explain why you revised your goals or assessment plan. If you didn't change anything, just write "No changes from phase 1" at the top of the section. Due date: N/A. This is optional.
Phase 3
At the end of your YOGA document, add a new section called Phase 3 (this is due 5/4, the day after the final event). Phase 3 should contain a section devoted to each learning goal. The section should have the following subsections:
What is the goal? [This should already be there from previous phases]
What is the assessment plan? [This should already be there from previous phases]
Evidence. Provide evidence (or pointers to evidence) that are useful in assessing the learning goal. This could be samples of code from your repo, a link to a video, a description of an activity you did.
Interpretation of the evidence. In your words, how does the evidence link with your assessment plan?
Your grade for this goal. This does not need additional justification beyond what was stated in the first four points. I will do some calibration of these among different students, so don't worry too much about knowing what other students are putting for these.
Project Proposal
You should push a document to your project repo that addresses the following questions. Please note that while you only need one proposal per team, the highlighted question below needs to be filled out per person.
Who is on the team?
In a couple of paragraphs, describe the key ideas of your proposed project? What is your MVP? What are your stretch goals?
To the best of your current knowledge, what datasets will you use for your project? Are there any obstacles you foresee in terms of getting access to the data?
What are the most important new skills / techniques you will have to learn to be successful in this project? If you think some of these skills would be useful for us to cover in class, please indicate which ones.
Outline a rough timeline for the major milestones of your project. This will mainly be useful to refer back to as we move through the project.
What do you view as the biggest risks to you being successful on this project?
Given each of your YOGAs (see here), in what ways is this project well-aligned with these goals, and in what ways is it misaligned? If there are ways in which it is not well-aligned, please provide a potential strategy for bringing the project and your learning goals into better alignment. There should be an individual section for each person on the team addressing the fit between the YOGA and the project topic.
This is due 3/25.
Website / blog
You should create a project website. The project website will serve two purposes: to show off your final deliverable and to document the story of how you did the work. In some cases your final deliverable might actually be a website, in this case the deliverables in this section can be added to this website (rather than creating a second website with just the contents described below). You should create the deliverables in this section for multiple audiences: general readers interested in your project, potential employers, other students in the course, and me.
Project Stories (blog)
Twice during the project you will be writing a blog entry that tells a story about some phase of your project. Possible templates for these stories are:
We had an interesting question about our dataset. We tried a particular method to answer the question, here's what we discovered! (fill in with lots of good descriptive text, code snippets, and visuals).
We had a particularly nasty bug. We tried lots of methods to debug the problem. Finally we figured it out! We learned lessons for the future. Hurray! (fill in with a detailed description of what the bug was and what methods you tried).
We were interested in applying a new data analysis technique. At a high level here's how the technique works and why we are interested in applying it to our data. Here's an example of this new technique being applied to our dataset.
We have implemented a key feature. It works really well! Here's how we did it. Hurray!
Project story 1 will be due on 4/5.
Project story 2 will be due on 4/22.
Turn in project stories by placing a link in this spreadsheet
Project Documentation
At the end of the project your website should include the following information:
Show it off: Describe the main findings / outcomes of your project. The exact nature of this section will of course depend on your specific project. One way to think of this page is if someone reads your project homepage and is intrigued by what he or she finds, this page should be where the reader would go next for a more detailed view of the work that you did. The bottom line is if you want me to look at it to assess your project, it should be findable via the project website.
System architecture: In detail describe the main analysis techniques used in your project. This should should be complementary to the content in "show it off" in that it is more focused around the "how" of the project work rather than the "what".
Project documentation is due 5/4 (the day after the final event).
Poster / Demo
On 5/3 we will be having a data science expo. You should prepare a poster for this expo. The exact size is up to you, but as a rough guideline I wouldn't go any smaller than 36" by 36". Here are some guidelines:
The poster need not be standalone. You should design this poster to serve as an aid to presenting your work to a small group of people that will come by your poster at the data science expo. What this means is that you need not design your poster to be navigable without you being there. It's okay if it is, but not necessary.
Make sure the poster is well-organized and supports you telling the story of your project
Make sure the visual elements of the poster are effective (e.g. text is appropriately sized)
This is designed to borrow heavily from other project deliverables, if you find yourself spending an inordinate amount of time creating this poster than you should probably check in with me as this is not at all my intention.
Code (encompasses functionality, design, documentation)
This is pretty self-explanatory. I will be using the same rubric as for the other assignments. If your repo is hard to navigate, please use the README.md to help provide some guidance for me and other interested parties.