[{"data":1,"prerenderedAt":161},["ShallowReactive",2],{"project-36061427":3},{"id":4,"name":5,"line":6,"line_source":7,"title":8,"url":9,"domain":10,"maker":11,"posted_at":12,"year":13,"points":14,"comments":15,"rank_in_year":16,"shown_in_year":17,"top":18,"tier":19,"category":20,"subcategory":21,"format":22,"keywords":23,"reception":27,"sentiment":28,"stance_mix":29,"scored_comments":32,"aspects":33,"strip":35,"post_text":36,"what":37,"said":62,"maker_in_thread":104,"shape":109,"grid":113,"attempts":124,"same_week":125,"maker_shows":153,"maker_hidden":154,"reader_version":155,"card":156},"36061427","Yakread","An RSS reader powered by machine learning","title","Show HN: Yakread – An RSS reader powered by machine learning","https:\u002F\u002Fyakread.com\u002F","yakread.com","jacobobryant",1684949383,2023,18,5,1769,3109,57,"shown","productivity","rss reader","web_app",[21,24,25,26],"article recommendations","newsletter reader","bandit algorithm","quiet",null,{"positive":30,"neutral":30,"mixed":31,"critical":30},1,0,3,{"idea":34},{"positive":30,"critical":30},[32,15,31,31,30],"This is a web-based reading app I've been working on since August. The main differentiator is that Yakread uses machine learning to rank the articles in your feed: as you click on articles from a particular RSS\u002Fnewsletter subscription, other articles from that subscription will tend to be ranked higher in the future (via a bandit algorithm). Yakread also uses ML to recommend articles that other users have read, so your feed will have articles in it even before you sign up and add your own subscriptions.\n\nFor the recommendations, I'm using the collaborative filtering implementation from Spark M",{"problem":38,"audience":44,"pricing":28,"motivation":50,"time_to_build":28,"stack":54,"openness":59,"stage":60,"team_size":61},{"value":39,"source":40,"evidence":41},"Finding and reading long-form articles from subscriptions, with ML ranking and recommendations","maker",[42],{"id":43},"post",{"value":45,"source":40,"evidence":46},"Readers of long-form content, aiming at mainstream users beyond the HN demographic",[47],{"id":48,"by":11,"text":49},"36071840","The key question I think is \"how do you create a recommender system that serves the user's interests instead of just the business's interests?\" And how do you help the user have…",{"value":51,"source":40,"evidence":52},"Scratches a personal itch and fixes deficiencies of the maker's previous businesses",[53],{"id":43},[55],{"value":56,"source":40,"evidence":57},"Spark MLlib collaborative filtering",[58],{"id":43},"unknown","launched","solo",{"praise":63,"objections":72,"requests":93,"questions":94,"alternatives":99},[64],{"value":65,"source":66,"evidence":67},"Interest in RSS plus recommendations as a standard-based approach","commenter",[68],{"id":69,"by":70,"text":71},"36068997","derbOac","I'm happy to see this and might give it a try as I've often wanted RSS + recommendations. I like being tipped off to things I wouldn't have found on my own and like RSS as a…",[73,80,87],{"value":74,"source":66,"evidence":75},"Delegating reading choices to AI conflicts with RSS users' values",[76],{"id":77,"by":78,"text":79},"36068688","hsjqllzlfkf","> More ideologically, I think RSS is ready for a comeback Ideologically, I love RSS too. I think it's one of the great achievements of the Internet, alongside with torrents and…",{"value":81,"source":66,"evidence":82},"ML is unnecessary for the described feature",[83],{"id":84,"by":85,"text":86},"36070281","aspyct","> Yakread uses machine learning to rank the articles in your feed: as you click on articles from a particular RSS\u002Fnewsletter subscription, other articles from that subscription…",{"value":88,"source":66,"evidence":89},"Public usage data for training could be de-anonymized",[90],{"id":91,"by":11,"text":92},"36072025","I've thought about this as well. Privacy is actually one impediment I think. If you make everyone's raw usage data public (so that anyone can use it to train their own…",[],[95],{"value":96,"source":66,"evidence":97},"How to build a recommender that serves user interests rather than business interests",[98],{"id":77,"by":78,"text":79},[100],{"value":101,"source":66,"evidence":102},"Twitter and Facebook feeds",[103],{"id":77,"by":78,"text":79},{"comments":105,"answered":30,"key":106},2,[107,108],{"id":91,"text":92},{"id":48,"text":49},{"commenters":110,"max_depth":32,"maker_replies":105,"first_comment_mins":111,"lifespan_hours":112},4,969,22,[114,116,118,120,122],[77,31,32,115,78,31],"maker\u002Fneutral\u002Fidea",[69,31,15,117,70,31],"project\u002Fpositive\u002Fidea",[91,31,31,119,11,30],"commenter\u002Fneutral\u002Fprivacy",[48,31,31,121,11,30],"project\u002Fneutral\u002Fidea",[84,30,30,123,85,31],"project\u002Fcritical\u002Fidea",[],[126,131,136,141,145,149],{"id":127,"name":128,"points":129,"category":130},"36009774","YouTube Full Text Search",540,"dev_tools",{"id":132,"name":133,"points":134,"category":135},"36064639","Pixian",383,"ai",{"id":137,"name":138,"points":139,"category":140},"36047143","Open sourcing Harmonic",305,"consumer_social",{"id":142,"name":143,"points":144,"category":130},"36020717","Trogon",248,{"id":146,"name":147,"points":148,"category":130},"36020001","SpaceBadgers",199,{"id":150,"name":151,"points":152,"category":135},"36037419","Blotter",166,7,false,"v2",{"id":157,"name":5,"line":6,"ai":154,"cat":20,"year":13,"points":14,"comments":15,"rec":27,"s":28,"top":18,"kw":158,"maker":11,"domain":10,"tier":19,"attempts":31,"strip":159,"url":9,"mix":160,"praise":65,"objection":74,"sub":21},36061427,[21,24,25,26],"35001",[30,30,31,30],1794036636060]