The problem i've been struggling with is how i properly summarize. The slot behavior is useful because slot-wins have material impact of finding content i do, actually, want to consume. Reducing the slot-pulls to once a day puts more weight on something else doing the filtering. Meaning i fear missing things i care about.
I wonder if there's some type of non-ML pattern seeking implementation that will let you refine your filter to improve signal:noise. I say non-ML because i'm interested in non-trained, smaller locally hosted datasets.
Anyway, i digress. This is cool! I think i'll let this thing be my slot-machine for the next few days and see if i feel like i'm missing anything. Appreciate it!