Expand on Java project using persistent data structures and additional similarity metrics.

It requires two programs. and create an order-32 B-tree holding word frequencies for each of at least 100 sites. Use the file system to map sites to datafiles. Use a buffer cache to speed up IO. Also, Pre-categorize (and somehow store) records into 5 to 10 clusters using k-means, k-mediods, or a similar metric. This would need to extend the previous order to display a category(cluster) and most similar key from the above data structures. The existing program is a web page categorization program. The program reads 10 (or more) web pages. The urls for these web pages can be maintained in a control file that is read when the program starts. For each page, the program maintains frequencies of words. The user can enter any other URL, and the program reports which other known page is most closely related, using a similarity metric. It uses existing library collections for all data structures, except for a custom hash table class which is maintaining word frequencies. It establishes a similarity metric. Which is part based on word-frequencies. It uses jsoup as a parser to extract words or other components. And has a gui, made with javafx.