Palladio is a great tool for examining connections within a data set. You can visualize how people are connected to a school or workplace, how many books have a trait like stack location, copyright year, or publisher in common, and how many interviewers asked their subjects the same questions. As long as you’re comparing data sharing connections, the possibilities are endless.

To try Palladio, I recommend starting with their example data. Navigate to http://hdlab.stanford.edu/palladio/, and click the “Start>>” button in the middle of the page. On the left, click “Try with sample data.” The “Data” view of the sample data will then be displayed. You’ll see two white rectangles each representing a table, or set of data. The one on the left is labeled “People” and is the primary table. You can see that it contains 73 rows, or entries. Those entries are made up of fields, or elements of data. In this case, the fields are information about a person: their name, birthplace, birthdate, gender, etc.

The right table is labeled “Places,” and it has far less data, just a Place (a text name) and Coordinates (the longitude and latitude of the place). Palladio can link data in different tables. If you look at the People table, you’ll see that the Birthplace, Arrival Point, and Place of Death fields all have green text to the right saying they have a certain number of Places. Those Places are part of the Places table. Essentially, the People table only knows the names of the Places, but Palladio can look up that name in the Places table to convert it to lat-long coordinates that can be mapped.

Speaking of mapping, click “Map” at the top-left of the page. You’ll see a blank world map with no data, but you can add data in the “Map layers” box. Click “New layer” and select a kind of data from the “Places” dropdown. Click “Apply” to see that data mapped as a set of points. But Palladio is really useful for showing the connections within a dataset. So let’s map some connections. Edit the data you added to the map by clicking the pencil icon to the right of the Data/Shapes layer you added in the “Map layers” box. Select “Point to point” for the Map type. Then choose “Birthplace” as the Sources Places and “Place of Death” as the Target Places. This will show us connections between where people were born and where they died. Before you Apply these changes, click the “Size points” checkbox and then select “Number of People” in the “According to dropdown that appears. This option changes the display of the points on the map so that they will be bigger if they represent more people. Click Apply, and explore the map!

We can explore the same data as a graph, which lets us visualize more relationships than just geographic ones. Click “Graph” at the top-left of the page. The graph starts blank, but you can add data by clicking the menu button to the top-right. In the “Settings” menu, select “Birthplace” as the Source and “Place of Death” as the Target. You’ll immediately see the graph populate with those connections. Click the Highlight checkbox under Source. This makes the Birthplaces darker so you can tell the differences between where people were born and died. Just like on the map, click the “Size nodes” checkbox and make sure the “According to” dropdown is set to “Number of People.” This shows us how many people were born and died at a place via the size of the point representing the place.

If you want to limit the connections displayed in the graph, you can filter them using the “Facet” menu on the bottom-left of the graph. Click “Facet” and then select “Gender” from the Dimensions dropdown. You’ll see the genders in this dataset appear to the left. Try selecting M or F. This will change the graph to show only people with the gender you selected. You can add more Facets to filter the data in different ways. For now, let’s remove any Facets you’ve added by clicking the Xs next to each one.

You can also graph data that is not geographic. Return to the Settings menu at the top-right, and select Position for Source and Gender for Target. Now we have a graph of how the gender and positions of the people in our data are connected. You’ll see that most of the positions are connected to both the M and F circles, showing that both males and females had those positions. But other positions are only connected to either M or F representing that all the people with those positions had the same gender. For instance, in this data, all the gamblers were male, and all the spouses were female. Try exploring other connections in the dataset. If you create a graph you want to save, you can click the “Download” link on the Settings menu to save the graph as an image.

Palladio has lots of options to display and filter data. Beyond the Facet filter functionality, you can also see data as a timeline or a group of timespans. You can also view the data as a table in the Table view and see individual entries from the Primary Table in the Gallery view.