Show HN: FeedSense – 基于你自己的信息源构建的私有推荐系统

2作者: langtang19964 个月前
Hi HN, 我构建这个东西的灵感主要来自两点:Andrej Karpathy 最近谈到回归 RSS 订阅源——“更高质量的长篇内容,以及更少旨在挑衅的垃圾信息”(<a href="https:&#x2F;&#x2F;x.com&#x2F;karpathy&#x2F;status&#x2F;2018043254986703167" rel="nofollow">https:&#x2F;&#x2F;x.com&#x2F;karpathy&#x2F;status&#x2F;2018043254986703167</a>)。大约在同一时间,Marc Andreessen 分享了他的信息来源:25% 来自 X,25% 来自从业者的播客,25% 来自与 AI 的对话,25% 来自旧书(<a href="https:&#x2F;&#x2F;x.com&#x2F;pmarca&#x2F;status&#x2F;2030949039345467570" rel="nofollow">https:&#x2F;&#x2F;x.com&#x2F;pmarca&#x2F;status&#x2F;2030949039345467570</a>)。 这让我意识到控制你接收的信息有多么重要。但我已经习惯了feed流式的阅读习惯(每个社交媒体都是这样)。我只是不想让一个随机的梗图或煽动性的内容突然劫持我的注意力。 所以我构建了 FeedSense——一个私人的推荐系统。你选择你的信息源,它就会只从这些信息源构建一个“为你推荐”的feed流。该系统会考虑内容质量、发布时间和你的阅读历史(仍在调整中),但它永远不会在你的选择之外呈现任何内容。 iOS 上的 TestFlight beta:<a href="https:&#x2F;&#x2F;testflight.apple.com&#x2F;join&#x2F;56VPEeeS" rel="nofollow">https:&#x2F;&#x2F;testflight.apple.com&#x2F;join&#x2F;56VPEeeS</a> <a href="https:&#x2F;&#x2F;feedsense.cc" rel="nofollow">https:&#x2F;&#x2F;feedsense.cc</a> 如果有人感兴趣,我很乐意讨论架构或推荐方法。
查看原文
Hi HN,<p>Two things inspired me to build this: Andrej Karpathy recently talked about going back to RSS feeds — &quot;higher quality longform and a lot less slop intended to provoke&quot; (<a href="https:&#x2F;&#x2F;x.com&#x2F;karpathy&#x2F;status&#x2F;2018043254986703167" rel="nofollow">https:&#x2F;&#x2F;x.com&#x2F;karpathy&#x2F;status&#x2F;2018043254986703167</a>). Around the same time, Marc Andreessen shared his information source: 25% X, 25% podcasts with practitioners, 25% conversations with AI, 25% old books (<a href="https:&#x2F;&#x2F;x.com&#x2F;pmarca&#x2F;status&#x2F;2030949039345467570" rel="nofollow">https:&#x2F;&#x2F;x.com&#x2F;pmarca&#x2F;status&#x2F;2030949039345467570</a>).<p>This made me realize how much it matters to control what you let in. But I&#x27;ve used to the feed-style reading habit (each social media does like this). I just don&#x27;t want a random meme or rage bait to hijack my attention out of nowhere.<p>So I built FeedSense — a private recommendation system. You pick your sources, and it builds a &quot;For You&quot; feed from only those sources. The system takes content quality, publish time, and your read history into account (still tuning it), but it will never surface anything outside the sources you chose<p>TestFlight beta on iOS: <a href="https:&#x2F;&#x2F;testflight.apple.com&#x2F;join&#x2F;56VPEeeS" rel="nofollow">https:&#x2F;&#x2F;testflight.apple.com&#x2F;join&#x2F;56VPEeeS</a><p><a href="https:&#x2F;&#x2F;feedsense.cc" rel="nofollow">https:&#x2F;&#x2F;feedsense.cc</a><p>Happy to talk architecture or the recommendation approach if anyone&#x27;s curious.