Heat Maps

This is cool charts #4.

A heat map (sometimes spelled as a single word) is a two dimensional chart in which areas of the chart display some sort of color scale mapped to a scale level variable. The name is probably derived from the displays on thermal imaging devices which are used to look for cold and hot areas.

The dimensions defining the axes can be almost anything. They are usually scalar since categorical values are arbitrarily ordered. That is, change the order of categories and the heat map is going to look very different with the same underlying data.

Heat maps are heavily used in business analytics, layout optimization, and capacity planning.

When the axes are longitude and latitude, the heat map is literally a map and is often overlaid on a geographic map. Here are two such examples.

New York Taxi Map

Source: New York Times

This points are the average number of pick-ups at a location per hour. Click on the link to go to the full map and move the time slider to see where’s the best place to find a cab at a particular time and day. Notice that the weekday pattern is really different from the weekend pattern as you might expect. If you hover over a point you’ll see the underlying data value for that point.

No Swearing in Utah

Source: Daniel Huffman's Cartostrophe Blog via Flowing Data

This is a heat map of the average number of tweets containing swear words by their location. The lighter the hue, the more frequent the swearing. Notice that Utah is significantly darker than Colorado and somebody must have been having a really bad day in the Oklahoma panhandle. The completely black area is the Great Salt Lake. Apparently, nobody tweets from the middle of it.

One of the major problems with a heat map is that it doesn’t so much plot a process as reflect underlying population differences. This map tries to avoid that in two ways: it uses an average of scatological tweets instead of an absolute number; and is uses isolines to define areas in which 500 tweets occur. The bigger the blob, the smaller the tweeting density.

Click on the map to see a big map of the US and read all the comments. This chart was also featured in Flowing Data. Go there to see other comments.


About ViAnn

Progressive retired geek who loves to play golf
This entry was posted in Tech. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s