I could imagine, as John Lennon used to sing, that "I am not the only one" in need of such an application.
At times you get ten of thousand lines which you would like to quickly “visually parse” to gain a general sense of what you've got. Ideally, you should be able to play with it to select the records you need.
Think for example, of the many links to texts you would get from archive.org (which also includes some metadata) or *.pub (each site using their own quirkiness)
Based on some sort of GUI, you would see weighted terms (coloured or not based on a user's preference) with all other terms preceding (as some sort of tree-like structure confluent on that term) and following it ( ... branching off of it).
Which kind of applications people use to do such thing?
lbrtchx