[{"data":1,"prerenderedAt":195},["ShallowReactive",2],{"project-3616763":3},{"id":4,"name":5,"line":6,"line_source":7,"title":8,"url":9,"domain":5,"maker":10,"posted_at":11,"year":12,"points":13,"comments":14,"rank_in_year":15,"shown_in_year":16,"top":17,"tier":18,"category":19,"subcategory":20,"format":21,"keywords":22,"reception":28,"sentiment":29,"stance_mix":30,"scored_comments":34,"aspects":35,"strip":38,"post_text":39,"what":40,"said":70,"maker_in_thread":132,"shape":134,"grid":137,"attempts":158,"same_week":159,"maker_shows":32,"maker_hidden":187,"reader_version":188,"card":189},"3616763","rogerbraun.net","A client-side Bayes classifier for Hacker News","title","Show HN: A client-side Bayes classifier for Hacker News","http:\u002F\u002Frogerbraun.net\u002Fa-client-side-bayes-classifier-for-hacker-new","rogerbraun",1329838065,2012,119,10,171,1677,11,"notable","productivity","news filtering","web_app",[23,24,25,26,27],"bayes classifier","hacker news","client-side","story filter","javascript","mixed",0.286,{"positive":31,"neutral":32,"mixed":32,"critical":33},3,1,0,5,{"design":36,"idea":37},{"positive":33,"critical":33},{"positive":31,"critical":33},[33,33,31,31,33,34,33,33,34,34],null,{"problem":41,"audience":53,"pricing":39,"motivation":39,"time_to_build":39,"stack":57,"openness":69,"stage":69,"team_size":69},{"value":42,"source":43,"evidence":44},"Too many Hacker News stories to read; filtering to ones the reader likes","inferred",[45,49],{"id":46,"by":47,"text":48},"3618276","gauravk92","Maybe it's easier simply to classify things you wouldn't want to read and hide those as less interesting. Because of the variety of topics, training something to figure out what…",{"id":50,"by":51,"text":52},"3618311","achompas","HN's algorithm is great at pushing linkbait, self-help articles, and frameworks associated with popular languages to the top. None of those are interesting to me, however, so a…",{"value":54,"source":43,"evidence":55},"Hacker News readers",[56],{"id":50,"by":51,"text":52},[58],{"value":59,"source":43,"evidence":60},"JavaScript",[61,65],{"id":62,"by":63,"text":64},"3617179","polyfractal","Very cool! I've been hacking around with modifying HN's interface via JS a lot recently - this will be a welcome tool in my experiments. One comment: The up\u002Fdown votes are really…",{"id":66,"by":67,"text":68},"3618057","noelwelsh","Agreed. I love the idea of implementing it in browser. So simple and elegant, and obvious in hindsight. Why didn't I think of it!?","unknown",{"praise":71,"objections":88,"requests":104,"questions":109,"alternatives":120},[72,77,84],{"value":73,"source":74,"evidence":75},"Simple and elegant, obvious in hindsight to do in the browser","commenter",[76],{"id":66,"by":67,"text":68},{"value":78,"source":74,"evidence":79},"Interesting starting point for experiments",[80],{"id":81,"by":82,"text":83},"3616897","growt","Nice work. I hope it gets more attention in the next hours. Seems like an interesting starting point for all kinds of experiments.",{"value":85,"source":74,"evidence":86},"Useful for hacking on HN's interface with JS",[87],{"id":62,"by":63,"text":64},[89,96,100],{"value":90,"source":74,"evidence":91},"Bayes classifiers reached only about 60% accuracy in one commenter's tests",[92],{"id":93,"by":94,"text":95},"3617972","moconnor","I tried to train bayesian (and other) classifiers to reliably pick the same stories to read as I would. Despite looking at a variety of things - title, poster, domain, corpus…",{"value":97,"source":74,"evidence":98},"Training on liked stories is restrictive; classifying uninteresting ones may work better",[99],{"id":46,"by":47,"text":48},{"value":101,"source":74,"evidence":102},"Upvote\u002Fdownvote controls are visually too strong",[103],{"id":62,"by":63,"text":64},[105],{"value":106,"source":74,"evidence":107},"Make the up\u002Fdown vote buttons smaller or lighter",[108],{"id":62,"by":63,"text":64},[110],{"value":111,"source":74,"evidence":112},"Whether the HN front page ranking is already good enough",[113,117],{"id":114,"by":115,"text":116},"3618258","Gring","As an alternative, just trust the HN home page algorithm. Stories seem to move up to a relevant max rank position, stay there and then move back down. Big stories stay in the top…",{"id":118,"by":115,"text":119},"3618470","Don't you think that submissions are motivated by getting to the home page? So with time, submissions become similar to what's on the home page. If the home page is so bad…",[121,125],{"value":122,"source":74,"evidence":123},"Relying on the Hacker News home page ranking",[124],{"id":114,"by":115,"text":116},{"value":126,"source":74,"evidence":127},"Manual curation with regex filters and votes for a newsletter",[128],{"id":129,"by":130,"text":131},"3618134","duck","Yeah, I think it shows how hard it is to classify something with such a diverse set of stories. Each week for my Hacker Newsletter project I have to come up with a short list of…",{"comments":33,"answered":33,"key":133},[],{"commenters":135,"max_depth":31,"maker_replies":33,"first_comment_mins":136,"lifespan_hours":14},9,30,[138,140,142,144,147,149,151,152,155,157],[93,33,33,139,94,33],"commenter\u002Fneutral\u002Fidea",[129,33,33,141,130,33],"commenter\u002Fneutral\u002Fnone",[62,32,31,143,63,33],"project\u002Fmixed\u002Fdesign",[46,145,31,146,47,33],2,"project\u002Fneutral\u002Fidea",[114,31,33,148,115,33],"other\u002Fneutral\u002Fnone",[50,31,34,150,51,33],"project\u002Fpositive\u002Fidea",[118,31,33,139,115,33],[153,31,33,148,154,33],"3618892","tikhonj",[81,156,34,150,82,33],4,[66,156,34,150,67,33],[],[160,165,170,174,178,183],{"id":161,"name":162,"points":163,"category":164},"3616603","CloudShark",164,"security_privacy",{"id":166,"name":167,"points":168,"category":169},"3600582","Photo Picker+",121,"dev_tools",{"id":171,"name":172,"points":13,"category":173},"3610927","QR Pixel","creative",{"id":175,"name":176,"points":177,"category":173},"3602160","Mothereffinanimatedgif",113,{"id":179,"name":180,"points":181,"category":182},"3608725","recommendify",97,"data",{"id":184,"name":185,"points":186,"category":169},"3608528","StyleBootstrap",88,false,"v2",{"id":190,"name":5,"line":6,"ai":187,"cat":19,"year":12,"points":13,"comments":14,"rec":28,"s":191,"top":17,"kw":192,"maker":10,"domain":5,"tier":18,"attempts":33,"strip":193,"url":9,"mix":194,"praise":73,"objection":90,"sub":20},3616763,0.29,[23,24,25,26,27],"0033050055",[31,32,32,33],1793990879048]