Cluster Analysis and Wheel Visualization.

Project Description

Building a series of breakthrough visualizations for many analysis tasks on the DOMO platform. You areseeking a qualitatively improved way to view clusters of information, compared to existing methods. Viewing data that naturally “clusters together” is of value in many application domains, including data formatted as surveys,transactions, and text. In preparation for this project, we collaborated on a UI sketch, which we have rendered in a mockup image below.

Key elements include the following:

1) Data for the cluster analysis is derived by a similar process as we followed in Market Basket. A transaction file is converted to a cluster file by our back-end AWS framework, which is then returned to the front end.

The difference is that, here, clusters may contain more than the two items. Some details:

a. As with the Market Basket application, we will create an iFrame within the DOMO application that the user can use to send transaction data to a back-end processor, which does combinatoric search to find the clusters with greatest support. Note that, as with Market Basket, this back-end task is nontrivial and if configured incorrectly can take a very long time.

b. To generate the clusters, we will use the frequent itemset method12. This approach uses heuristics to reduce the time required for the 2^n search for all clusters; the algorithm involves