Working with complex information is part of the scientific process and can lead to valuable findings and new directions of research. But data can be intimidating to someone without scientific training. If we want to engage the public in science, as well as fit our research into a broader societal framework, we must be able to use our data to tell a story.
Data visualization is one way to enable an audience to extract important content from large amounts of information. When used properly, it can effectively bridge the gap between technical results and useful interpretations. In short, it can help you answer the question: “Why should I care?"
I recently learned about Gephi, a user-friendly, freely available software which generates visualizations from user-input data. Most usefully, it features a modularity function, which sorts the data into communities. In other words, it will show you clustered groups based on their connections with each other by comparing it to the connections that might exist due to random chance.
For example, I can import my Facebook network data (downloaded using NetGet) and use Gephi to visualize all of my Facebook friends and how they are connected. The modularity function will automatically cluster and color-code communities, which I can then easily identify.
If you look at the visualization below, you can see circles (nodes) representing my Facebook friends (I have removed any identifiers) and lines (edges) connecting the circles if they are friends with each other. You can see that that the orange, green, and red communities represent different networks from my hometown of Atlanta. The orange and green communities share quite a bit of overlap because many people that I know from club swimming also went to my high school. Additionally, my college and graduate school networks are more isolated because they are in different geographical locations and therefore less likely to share connections.
With a visualization like this, it is also interesting to explore the people who bridge more isolated communities. For example, I know several people who are connected to both my club swimming and college (where I also swam) networks. In this case, data visualization allowed me to easily identify specific people or communities without having to search through my Facebook friends manually.
Social networks are just one type of data to explore. Gephi allows you to import any information, which can be organized as nodes or edges, as well as map other features such as the weight, or strength, of connections. One could imagine looking at protein-protein interactions, neural connections, or even co-authorships across papers.
Gephi is one of many widely available tools to visualize complex information. It can serve as a useful way to gain a quick impression of your data, as well as aid in presentation displays or figures. While it may take some time to learn how to best visualize your data, it will certainly prove useful in communicating a more effective and impactful message.