Generating Network Analysis Insights in Alteryx and Tableau

Posted by Anusheela Banik

This is the followup to How Network Analysis Can Add Value. Aimpoint Digital wanted to share their experience in producing these graphs in Alteryx and Tableau. Although the technique is not technically native to either, Alteryx’s predictive toolkit allows us to visualize the data in both tools.

Prep the data

Source data for this analysis is updated daily via PDF reports. This data was compiled, cleansed, and enhanced using various tools within Alteryx to create a single database of known cases through time. To use Alteryx’s native network analysis capabilities, the data must include a unique set of nodes, for which each distinct case was one. The contact tracing efforts of Singapore’s Ministry of Health provided the necessary information to create labeled links, or connections, between the nodes. Using the make group tool and pivoting and unpivoting the data, the connections between cases were easily labeled for next steps.

Visualize

With the inputs complete, the R-based Network Analysis macro was dragged onto the canvas with settings tailored to Aimpoint Digital’s analysis. Alteryx uses D3.js to provide dynamic visualizations, but there was no inherent way to start to look for insights hidden within subsets of the data. Using the Reingold–Tilford algorithm for location and spacing within Alteryx to generate the graph, the coordinates for both nodes and edges were extracted to move the visualization into Tableau. By augmenting our main database of demographic information about each known case with a relative spatial location, we could import our network analysis graph into Tableau. By adding dynamic highlighters, a native feature of Tableau, certain elements of interest can be emphasized across the known cases. Additionally, with the data structured by date, this type of graph showed how the known cases spread over time and the arrival of new clusters. The resulting graph clearly demonstrated the spread over time and the importance of practicing social distancing. 

Curious about the data?

The data in this analysis was provided by the Singapore Ministry of Health (MoH). The MoH has been working on patient tracing and gathering demographic information, such as nationality, age, and gender, to understand how the virus has spread throughout their country. They notate links between cases as either family members or contacts. As the disease has spread, natural clusters of confirmed cases have been identified, all of which makes this analysis possible.

 

Aimpoint Digital will help you take an idea from thought through execution. This collaborative journey will enable you to get the most out of your data and technology investments. Contact us to begin your acceleration process.

Featured

Close
Contact Us