Inside-Out Additive Effects

The new Pixar film Inside-Out has emotion-characters that inhabit the mind of an 11-year old girl, named Riley. All five of her selected emotion-characters are important, Anger, Fear, Disgust, Sadness, and Joy, and as the film progresses they combine together to create new emotions. Christopher Haubursin from Vox created the matrix above, showing the new emotions created. For example Joy and Anger, either top-right or bottom-left of the matrix, combine together to produce Righteousness. Riley has many more emotions to choose from with the resulting additive effects. A clever display.

Correlation Ellipse Matrix

Here is an informative graphical representation of the correlation matrix of a data set of weather data on 16 variables. The image is via the RevolutionAnalytics blog and their post on big data available in R. The data mining R package Rattle produced the image. The magnitudes of correlations are shown with a concentration ellipses. Blue ellipses, with positive slopes, display variables with positive correlation with darker blue shading depicting higher positive correlation and lighter blue shading depicting lower positive correlation. Near zero correlations are shown as open, unfilled ellipses that are nearly circular. Variables with negative correlation are shown similarly by red ellipses with negative slopes. Dark red shading depicts greater negative correlation and lighter red shading depicts negative correlations closer to zero. This is a very useful tool to guide the eye through the relationships of the many variables used. Of course, as is well known, correlation is only an appropriate measure of the strength of the linear relationships between two variables when their scatterplot shows approximately the elliptical shape shown in these shaded icons. Examples abound of scatterplots where representing them as above can be misleading. My favorite, below, from the book by Chambers, et al. shows eight scatterplots all with the same positive correlation of 0.7. Correlation and its representation with an ellipsoidal icon is appropriate for only one of these scatterplots. This extols us all to remember, to look at the data.

Using Metadata to find Colonial Traitors to the English Crown (c. 1770s)

Last week Duke sociologist Kieran Healy had a fun and informative post showing how much information about social networks can be gleaned from so little metadata. Metadata have been in the news lately with the disclosure of Prism the US Government's communication surveillance system. Healy's post shows how knowing only what groups individuals belong to and a little matrix arithmetic that individuals can be "linked through the groups they belong to [and] groups can be linked by the people they share". Using data from the book "Paul Revere's Ride" by Fischer, Healy produced the basic image above showing the links between various Colonials through the clubs, committees, caucuses, etc that they belonged to. In particular this plot shows people that serve as bridges between the groups. I've added the portrait of Paul Revere, a central bridge, "ready to ride and spread the alarm", to many political groups and marking him as a notable Colonial traitor to the British Crown of the 1770s. Healy also has a plot showing the links between the groups. So much information from so little metadata. We've used other similarity plotting techniques here to cluster the Justices of the Supreme Court.

(The follow-up to Gym Freqs, promised last week, will appear next week).