score = lg(vote) - c*time
一个帖子的得分,与赞同数的对数正相关,与创建时间负相关。
各个网站有变种,比如著名的网站reddit的帖子排序
reddit score = lg(| agree - against |) + (t * y)/45000
(agree > against 时 y =1 ,agree < against 时 y =-1)
一个帖子的得分,与赞同数的对数正相关,与创建时间负相关。
各个网站有变种,比如著名的网站reddit的帖子排序
(agree > against 时 y =1 ,agree < against 时 y =-1)
本文标题:投票排名的基础模型
本文链接:https://www.haomeiwen.com/subject/byuopttx.html
网友评论