Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with
or
.
Download ZIP
Branch: master

Fetching latest commit…

Cannot retrieve the latest commit at this time

..
Failed to load latest commit information.
submissions
README.md
interestpoints.csv

README.md

Lab 8: Visualizing Data

Assigned: Tuesday, Nov 5, 2013

Due: Tuesday, Nov 12, 2013, 12:59PM (just before class)

The goal of this lab is for you to create a web-based visualization in D3 and learn something interesting about a local dataset.

The submissions are hosted as a gallery here!

http://www.mit.edu/~eugenewu/submissions/

Here are some useful documents

Datasets

In this lab, you will use a dataset of all taxi cab pickups in Boston proper over the course of one year.

Download the dataset here

Here are the first few lines of this file:

156527,2012-05-07 14:54:00,50 Francis St Boston Ma 02215,-71.107178,42.335685,126488
156528,2012-05-07 17:26:00,1080 Boylston St Boston Ma 02215,-71.08807,42.347138,126489
156529,2012-05-07 17:26:00,50 Seaport Ln Boston Ma 02210,-71.041327,42.349207,126490
156530,2012-05-07 14:55:00,Viaduct St Boston Ma,-71.041705,42.349905,126491
156532,2012-05-07 17:27:00,830 Boylston St Boston Ma 02199,-71.082857,42.348595,126492

The format is trip id,time,address,latitude,longitude,dropoff id. Here dropoff id is an identifier in a dropoffs data set (that we have not provided). Unfortunately, due to the way this dataset was generated, trip id is not guaranteed to be unique.

In the lab directory is also a file interestpoints.csv of interesting locations in Boston. Each line of this file is a named location and a latitude/longitude. These points were provided by the City of Boston as a list of locations where they are most interested in taxi activity.

Questions

Your goal for this lab is to construct a visualization that provides some insight into pattern of taxi rides through the interesting locations in Boston. You are free to display the data and locations however you want, but here are some things you may want to consider:

  1. Do you want to produce a static visualization or an animation (e.g., over time), or an interactive visualization that lets users vary some parameter?
  2. For the locations, you will probably want to define some radius around each location to count the number of pickups in that location.
  3. Think about the story you want to tell from the data. For example, you might try to visualize outliers in the dataset, or depict daily/weekly/monthly patterns, or show the differences in traffic between locations. The best visualization for each of these options is likely to be quite different.

Once you have made your visualization, answer the following questions:

  • What is the message in your visualization and what techniques did you use to illustrate it?
  • If you used interaction or animation, how does it facilitate the user's understanding?
  • What format is the data that is used by the visualization? Is it the raw data or did you need to compute some summary or derived data product?

Submission Instructions

Put your visualization and answer to the questions somewhere on the internet where we can access. http://bl.ocks.org/ is a nice place if you don't want to host yourself.

Add your URL and name to this spreadsheet. If you don't have a Google Drive account, feel free to email us your URL at 6885-staff@mit.edu.

Have you terminated your instances from previous labs?

aws ec2 terminate-instances --instance-ids <INSTANCE_ID>

Now you're almost done! Go read the assigned paper(s) for today.

You can always feel free to contact us with questions on Piazza.

Something went wrong with that request. Please try again.