什么是最好的算法来计算最砍下项目? [英] What is the best algorithm to calculate the most scored item?

查看:113
本文介绍了什么是最好的算法来计算最砍下项目?的处理方法,对大家解决问题具有一定的参考价值,需要的朋友们下面随着小编来一起学习吧!

问题描述

我都进球了,1间用户5,一个音乐项目, 我需要一个公式来获得最高的5计分项目。

I have an music items that are scored by users between 1 to 5, and I need a formula to get the 5 most scored items.

但显然是得到3.5平均得分从1000不同用户的项目将被更多的得分,然后一个项目泰德得到4.9平均得分仅5个用户......换句话说,我认为,如果一个项目获得关注的人得分它,这表示该项目是有趣。因此在计算中votesCount参数需要有一个电源。 (多大的权力?我不知道,我问它,你得到的想法)。

But obviously an item that get 3.5 average score from 1000 different users will be more scored then an item thet get 4.9 average score from only 5 users... in other words I think that if an item get attention from people to score it, this indicates that the item is interesting. so in the calculation the votesCount parameter need to have a power. (how much power? I don't sure, and I asking it you to get ideas).

我认为,我们需要以下参数功能:votesAverage,votesCount。

I think that we need the following parameters in the function: votesAverage, votesCount.

推荐答案

您可以使用贝叶斯估计来计算加权投票。

Weighted voting for 5-star systems with lots of voters

You can use Bayesian estimates to calculate weighted voting.

IMDB (互联网电影数据库)使用此计算,以确定其IMDB前250的(注:IMDB使用了10颗,但该公式是相同的使用5星)。

IMDb (Internet Movie Database) uses this calculation to determine its IMDb Top 250. (Note: IMDb uses 10 stars but the formulas are identical using 5 stars).

该公式计算评分最高250标题给人以真正的贝叶斯估计:

The formula for calculating the Top Rated 250 Titles gives a true Bayesian estimate:

加权评级(WR)=(V÷(V + M))×R +(M÷(V + M))×C

weighted rating (WR) = (v ÷ (v+m)) × R + (m ÷ (v+m)) × C

其中:

      
  • 在R =平均影片(平均)=(额定值)
  •   
  • V =票的电影数量=(票)
  •   
  • M =所需的最低票数列在250强(目前3000)
  •   
  • C =在整个报告中的平均投票(目前为6.9)
  •   
  • R = average for the movie (mean) = (Rating)
  • v = number of votes for the movie = (votes)
  • m = minimum votes required to be listed in the Top 250 (currently 3000)
  • C = the mean vote across the whole report (currently 6.9)

IMDB参考

维基百科参考

这篇关于什么是最好的算法来计算最砍下项目?的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持IT屋!

查看全文
登录 关闭
扫码关注1秒登录
发送“验证码”获取 | 15天全站免登陆