Contextual word cloud

Hi All,

I am trying to build a contextual word cloud. Basically, a set of two word clouds where you can select a term in the first visual and the second visual will filter to show all of the terms used in phrases that contain the term selected in the first visual. For instance, we have been working with some open survey responses passed through Azure's key phrase tagging API. One of the most mentioned terms is 'Lunch'. To give me some additional information, I would like to see the terms used in any phrase that also mentions 'Lunch'.

Currently, my dataset contains three columns: A uniqueID (unique to a respondent), the full text string, and the key phrase tags.

I have had some success in duplicating the dataset and joining on the ID. If I filter one dataset by selecting a term in the first word clound it will filter the second word cloud by ID. The issue is that the selected term is the most prevalent in the second word cloud and makes it hard to review the contextual terms.

I am happy to provide additional detail. Has anyone had success with this type of process? Thanks!