This document contains guidelines, requirements, and suggestions for Project 1.
Before anything, remember that Projects are a group effort: Working closely with your teammates is a requirement. This both helps teach real-world collaborative workflows, and enables you to tackle more difficult problems than you'd be able to working alone.
In other words, working in groups allows you to work smart and dream big. Take advantage of it!
Before you start writing any code, your group should outline the scope and purpose of your project. This helps provide direction and prevent scope creep.
Write this as a brief summary of your interests and intent, including:
-
The kind of data you'd like to work with/field you're interested in (e.g., geodata, weather data, etc.)
-
The kinds of questions you'll be asking of that data
-
Possible source for such data
In other words, write down what kind of data you plan to work with, and what kinds of questions you'd like to ask of it. This constitutes your Project Proposal/Outline, and should look something like this:
Our project is to uncover patterns in criminal activity around Los Angeles. We'll examine relationships between types of crime and location; crime rates and times of day; trends in crime rates over the course of the year; and related questions, as the data admits.
Once your group has written an outline, it's time to start hunting for data. You are free to use data from any source, but we recommend the following curated sources of high-quality data:
Chances are you'll have to update your Project Outline as you explore the available data. This is fine—adjustments like this are part of the process! Just make sure everyone in the group is up-to-speed on the goals of the project as you make changes.
Make sure that your data is not too large for local analysis. Big Data datasets are difficult to manage locally, so consider a subset of that data or a different dataset altogether.
With data in hand, it's time to tackle development and analysis. This is where the fun starts!
Inevitably, the analysis process can be broken into two broad phases: Exploration & Cleanup and Analysis proper.
As you've learned, you'll need to explore, clean, and reformat your data before you can begin to answer your research questions. We recommend keeping track of these exploration and cleanup steps in a dedicated Jupyter Notebook, both for organization's sake and to make it easier to present your work later.
Similarly, after you've massaged your data and are ready to start crunching numbers, you should keep track of your work in a Jupyter Notebook dedicated specifically to analysis.
During both phases, don't forget to include plots! Don't make the mistake of waiting to build figures until you're preparing your presentation. Creating them along the way can reveal insights and interesting trends in the data that you might not notice otherwise.
We recommend focusing your analysis on techniques such as aggregation, correlation, comparison, summary statistics, sentiment analysis, and time series analysis.
Finally, be sure that your projects meet the technical requirements.
After you've analyzed your data to your satisfaction, you'll put together a presentation to show off your work, explain your process, and discuss your conclusions.
This presentation will be delivered as a slideshow, and should give your classmates and instructional staff an overview of your work. PowerPoint, Keynote, and Google Slides are all acceptable for building slides.
As long as your slides meet the presentation requirements, you are free to structure the presentation however you wish, but students are often successful with the format laid out in the presentation guidelines.
Coding Boot Camp © 2017. All Rights Reserved.