Submission:
- Please push your assignment to your fork (your GitHub repository of the course) and submit a link to it via the form shared in Slack.
- You will pitch your ideas in class in small groups, get feedback for your own final project and provide feedback for others as well.
In the field of data science, projects are practical. A good project is manageable and relates to your working domain; however, it can be hard to filter and scope good ideas when you're new to a field. That's where Part 1 of your final project comes in.
One of the best ways to test expectations and get feedback is to share your ideas with others. For Part 1 of your final project, you'll come up with a few different ideas you could potentially solve with data, then present them in small groups to the class.
You'll get the opportunity to get immediate feedback and guidance in order to help you choose an awesome final project idea. Keep in mind that it's important to run though this ideation process a few times with different ideas to clarify what your project should focus on.
Objective: Prepare a two- to three-minutes lightning talk covering three potential project topics.
-
Requirements: You should present from a prepared deck that explains the following:
- The Problem: What's the background and scope of the project idea? What problem are you attempting to address or solve? Who may it matter to?
- The Data: What data exists to help solve this problem? Where is it coming from? What does the data look like? What is the observation?
- Your Hypotheses: Given the problem and data you're aware of, what do you believe is the solution? What does success look like?
-
Detailed Breakdown: Your presentation must:
- Explain three potential projects, including: a problem statement, a hypothesis, and potential datasets.
- Be two to three minutes in duration.
- Demonstrate familiarity with the domain of the data.
- Listen to your peer's pitches and provide written feedback. Specifically:
- How might they iterate the hypothesis? What are alternative hypotheses?
- How might any risks/challenges be addressed?
- Try putting together a three-slide deck for each of your project ideas, in order to stay organized.
- Look for example (or real) dataset to get a feel for what information exists that you can use.
- Think about how much time you can dedicate to the project each week, and where you think the largest challenges will lie.
- Think about this as an elevator pitch: be prepared, do some light research, but don't go down any "rabbit holes".
- For overall project ideas, consider looking at former General Assembly Data Science Projects.