Comparing Digital Tools: Voyant, CartoDB and Palladio

31 Comments / Digital Humanities / By admin

There are a variety of tools that are available for doing digital history. In my last three blog posts, I have used Voyant, CartoDB and Palladio to analyze the Slave Narratives of interviews conducted in Alabama. In this blog post, I will compare these tools and conclude by showing how they might be more useful together.

These three tools: Voyant, CartoDB and Palladio have their different strengths and they each allow us to do the things that the others may not be able to. Voyant allows you to extract words from a corpus, see how often the words are used and correlate them with other words. Neither CartoDB nor Palladio are able to mine words. However, you can correlate them through maps in CartoDB and through network graphs in Palladio. The word cloud visualization is a strength in Voyant that one does not see in the other two.

The first image shows word cloud visualization of the interviews and the second image shows the number of times the word “Mother” was used in the different states.

CartoDB is excellent in map creations. Voyant lacks the ability to create maps. Palladio has mapping functions but its strength is not in mapping. The map functionality in Palladio is primarily for the visualization of relationships/ networks. CartoDB allows you to create different kinds of maps, export them or publish them. Unlike Palladio, you can access the information that is embedded in the different points that are on the map. CartoDB allows you to create different kinds of maps such as cluster maps, point maps, heat maps, etc.

CartoDB maps showing the different points where interviews took place. You can click on any of those orange dots to access the information.

Unlike Voyant and CartoDB, Palladio is excellent in the visualization of networks/relationships. Voyant and CartoDB do some of this but not in the detailed and advanced way that Palladio does. For example, Palladio is able to show us in graph the relationships in the data between the former slaves who are 90 and 91 years and the topics they talked about with their interviewer. This is not a capable you see in Voyant and CartoDB. The visualization of those relationships is the key with Palladio and that kind of visualization makes us understand the dataset better.

A graph of the Relationships of former slaves who were 90 and 91 years and the topics in their interviews.

What may we learn from these three tools? No single digital tool is able to extrapolate and analyze all data available. Some data may require different tools and sometimes you may need a combination of tools to do the kind of analysis that you want to do. In the case of these Slaves Narratives, all three tools played different kinds of roles to allow me to understand this vital data. I used Voyant to create word clouds and to graph the words used in the interviews; I used CartoDB to create maps that helped me visualize the data points where the interviews were conducted; and I used Palladio to link the relationships and topics in the interviews.