Ask HN: Best way to automate link categorizing / insights?

Over the past few months I've been working on putting together a little SQLite database that has in it a table containing every bookmark I've ever saved (browser bookmarks, YouTube playlists, Spotify playlists, Pocket article, Twitter bookmarks, Reddit saved, etc).

My table schema is essentially the date the bookmark was added, the URL of the bookmark, the title of the page at the time of bookmarking, and the source the bookmark came from (i.e. Chrome, Firefox, Reddit, etc). Of course I can always add more columns by scraping the URL in every row.

I currently have about 21,000 links aggregated into this database. My big question: What next? What should I do with this data? I've already built out full text search, aggregation by timestamp, search via raw SQL query, and some more pages. But I want to put the ball in your court: given this dataset, what could you tell about me? Conversely, what can I learn about myself by analyzing this table?