https://hackerpaper.kaipfister.de/p/1411.1127/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMThttps://news.ycombinator.com/item?id=38811967urn:hackerpapers:mention:1411.1127:38811967Sat, 30 Dec 2023 01:04:22 GMT<p>https://arxiv.org/abs/1411.1127</p><p><a href="https://hackerpaper.kaipfister.de/p/1411.1127/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.GThttps://news.ycombinator.com/item?id=13360880urn:hackerpapers:mention:1411.1127:13360880Mon, 09 Jan 2017 23:22:51 GMT<p>I think that the correct behavior in the long run is to give personalized recommendations, analogous to this proposal: https://sideways-view.com/2016/12/01/optimizing-the-news-fee... The main motivation for having a unified view is to decide what to show people who visit but don&#x27;t rate things / log in / etc. I&#x27;m currently finishing my dissertation on collaborative learning algorithms, and I agree that it is not generally possible to have a single global answer. This paper does something similar to your respect matrix: https://arxiv.org/abs/1411.1127 (The same technique can be modified to…</p><p><a href="https://hackerpaper.kaipfister.de/p/1411.1127/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.GT