Data Science: Capstone Overview
Capstone Project
Your Capstone project is the culmination of your time at GA. You are tasked with developing an interesting question, collecting the data required to model that data, developing the strongest model (or models) for prediction, and communicating those findings to other data scientists and non-technical individuals. This introductory document lays out the five consitutent portions of the project and their due dates.
Your Deliverables
The Capstone Project is divided into 5 deliverables, each building on top of skills learned previously to scaffold your learning over the entire course. Project deliverables include objectives, requirements, rubrics, and suggested resources - all of which tie into the overall competencies for this course.
Capstone, Part 1: Pitch + Problem Statement
Pitch us on potential ideas for a data-driven project. Think of topics you’re passionate about, knowledge you’re familiar with, or problems relevant to to industries you’d like to work with. What questions do you want to answer?
- Requirements: Lightning talk with 2-3 topics, including a problem statement, potential audience, goals, and success metrics, as well as possible data sources for each. Remember, if you can’t find data, you can’t do your project.
- Format: Slide deck
Capstone, Part 2: Dataset + Data Collection
Use your newfound skills to source and collect the relevant data for your project. Data acquisition, transformation, and cleaning are typically the most time-consuming parts of data science projects, so don’t procrastinate!
- Requirements: Source and format the data for your project. Perform preliminary data munging and cleaning of the data relevant to your project goals. Describe your data keeping the intended audience of your final report in mind.
- Format: Table, file, or database with relevant text file or notebook description.
Capstone, Part 3: EDA + Preliminary Analysis
Begin quantitatively describing and visualizing your data. With rich datasets, EDA can go down an endless number of roads. Maintain perspective on your goals and scope your EDA accordingly. Managing your own time is a critical skill in analysis projects. Keep notes on your approach, results, setbacks, and findings.
- Requirements: Perform initial descriptive and visual analysis of your data. Identify outliers, summarize risks and limitations, and describe how your EDA will inform your modeling decisions.
- Format: Jupyter Notebook
Capstone, Part 4: Findings + Technical Report
Share your technical findings with your fellow data scientists. Explain your goals, describe modeling choices, evaluate model performance, and discuss results. Data science reporting is technical, but don’t forget that you should tell a compelling story about your data.
- Requirements: Summarize your goals and metrics for success, variables of interest, and removal of any outliers or data imputation. Your process description should be concise and relevant to your goals. Summarize statistical analysis, including model selection, implementation, evaluation, and inference. Be convincing – justify all important decisions! Clearly label plots and visualizations. Include an Executive Summary.
- Format: Jupyter Notebook
Capstone, Part 5: Presentation + Non-Technical Summary
Take your findings and share a 15-minute, well-rehearsed presentation that delivers the most important insights from your project. Tell us the most interesting story about your data. Break down your process for a novice audience. Make sure to include compelling visuals. Time is short, so be sure to practice and include only the most relevant components of your project.
- Requirements: Convey your goals, limits/assumptions, methods and their justification, findings, and conclusions. Define technical terms. Include graphics and visualizations.
- Format: Public presentation of 13-17 minutes, plus 5 minutes of questions. Presentation should include interactive graphics, website, and/or slide deck.
Important note
Please ignore the _layouts folder and the config.yml file in the root of this repo. They are needed to make sure the repo renders well; please don’t change them or delete them and your repo will work fine :)