Effectively computing non-personalized recommendations can be annoyingly subtle. If we do naïve things like sorting by average rating, we get a top-N list dominated by items that one user really liked. Sorting by overall popularity doesn’t have this problem; as soon as we want contextual related-product recommendations, however (e.g. “people who bought this also bought”), and don’t want those recommendations to be dominated by the most popular items overall, the problem comes roaring back.

In my end-of-year blogging, I usually provide an update on my toolbox, and a review of the year.

So what’s happened? A quick (and likely incomplete) summary…

It’s that time of year again! Since I did a relatively full reset last year, for this year’s State of the Tools I’ll focus on changes and a few highlights. My Zoom studio post discusses my hardware and software for video recording in more detail.

Photo by Zalfa Imani on Unsplash

Inspired by Alan Said & Sole Pera’s posts on LinkedIn, here’s a quick summary of where you can find me and/or my work at RecSys next week in Bar, Italy:

There are more intervention strategies in heaven and earth, Horatio, than are dreamt of in your philosophy.
— Hamlet, sort-of

Tomorrow, I will be presenting our paper in the IR4Good track at ECIR 2024 on strategic IR impact interventions.

From my side, this paper grew out of a couple of things: