What is interpretable clustering and why is it important?
—
Clustering is a popular unsupervised learning task that groups similar data points. Despite this being a common machine learning task, most clustering algorithms do not explain the characteristics of each cluster or why a point identifies with a cluster, requiring users to do extensive cluster profiling. This time-consuming process becomes incredibly difficult as the datasets at hand grow larger and larger. This is ironic since one of the main uses for clustering is to discover trends and patterns in the data given.
With these considerations in mind, wouldn’t it be nice to have an approach that not only clustered the data but also provided innate profiles of each cluster? Well, that’s where the field of interpretable clustering comes into play. These approaches construct a model that maps points to clusters, and users are ideally able to analyze this model to figure out the qualities in each cluster. In this article, I am to discuss why this field is important as well as cover some of the main tracks of interpretable clustering.
If you’re interested in interpretable machine learning and other aspects of ethical AI, consider checking out some of my other articles and following me!