Question

我有三张桌子; articles，likes和shares。我想收集50篇文章，这些文章具有最多的个人观点，喜欢和分享。这将作为基于这三个条件获取趋势文章的查询。

我将在下面描述简化的表格结构：

制品

+----+-----------+-------+
| id | title     | views |
+----+-----------+-------+
| 1  | Article 1 |    92 |
| 2  | Article 2 |    14 |
| 3  | Article 3 |    39 |
| 4  | Article 4 |    87 |
| 5  | Article 5 |     8 |
+----+-----------+-------+

喜欢

+----+-----------+
| id | articleID |
+----+-----------+
| 1  |         2 |
| 2  |         3 |
| 3  |         2 |
| 4  |         5 |
| 5  |         3 |
| 6  |         3 |
+----+-----------+

股

+----+-----------+----------+
| id | articleID | type     |
+----+-----------+----------+
| 1  |         1 | facebook |
| 2  |         3 | facebook |
| 3  |         1 | twitter  |
| 4  |         4 | twitter  |
| 5  |         2 | facebook |
+----+-----------+----------+

我不确定获取热门文章的最佳公式。每篇文章的观看量总是占主导地位，这意味着如果我将总观看次数，喜欢和份额添加到一个汇总值，结果将或多或少取决于最高的观看量：

[0] => "Article 1"
[1] => "Article 4"
[2] => "Article 3"
[3] => "Article 2"
[4] => "Article 5"

我的问题是;如何制作最佳的趋势查询＆＃34;根据这三个条件？

更新

我正在寻找的公式是基于观看次数，喜欢和分享的整体百分比值。例如，第1条具有以下百分比值：

Views:  38.33% // Article's views divided by all articles' combined views (92 / 240)
Likes:  0%     // Article's likes divided by all articles' combined likes (0 / 6)
Shares: 40%    // Article's likes divided by all articles' combined shares (2 / 5)
Total:  78.33  // Calculation: 38.33 + 0 + 40

第1条将有＆＃34;趋势点＆＃34; 78.33。对所有文章执行此操作应提供以下结果：

[0] => "Article 3" // 86.25
[1] => "Article 1" // 78.33
[2] => "Article 2" // 59.16
[3] => "Article 4" // 56.25
[4] => "Article 5" // 19.99

如何进行此类MySQL查询？

Answer 1

视图始终是可预测的，因此我建议您跟踪多少百分比喜欢和分享观看次数。

Answer 2

我没有得到多少优化，但是要获得完全趋势输出，这就是我要做的：

create table articles (id integer, title varchar, views integer);
create table likes (id integer, articleID integer);
create table shares(id integer, articleID integer, type varchar);
insert into articles(id,title,views)values(1,'Article 1',92);
insert into articles(id,title,views)values(2,'Article 2',14);
insert into articles(id,title,views)values(3,'Article 3',39);
insert into articles(id,title,views)values(4,'Article 4',87);
insert into articles(id,title,views)values(5,'Article 5',8);
insert into likes(id,articleID) values(1,2);
insert into likes(id,articleID) values(2,3);
insert into likes(id,articleID) values(3,2);
insert into likes(id,articleID) values(4,5);
insert into likes(id,articleID) values(5,3);
insert into likes(id,articleID) values(6,3);
insert into shares(id, articleID, type)values(1,1,'facebook');
insert into shares(id, articleID, type)values(2,3,'facebook');
insert into shares(id, articleID, type)values(3,1,'twitter');
insert into shares(id, articleID, type)values(4,4,'twitter');
insert into shares(id, articleID, type)values(5,2,'facebook');

select A.id, A.title,COALESCE(
A.views+(select count(id) from likes where id=A.id group by id)
+(select count(*) from shares where articleid = A.id 
group by articleid),0) as Freq 
from articles A order by Freq desc;

 id |   title   | freq 
----+-----------+------
  1 | Article 1 |   95
  4 | Article 4 |   89
  3 | Article 3 |   41
  2 | Article 2 |   16
  5 | Article 5 |    0
(5 rows)

Answer 3

我可能已经使用下面的查询来解决这个问题：

SELECT      articles.id,
            articles.title,
            (
              coalesce(articles.views / A.totalViews, 0) +
              coalesce((SELECT count(*) FROM likes WHERE articleID = articles.id) / L.totalLikes, 0) +
              coalesce((SELECT count(*) FROM shares WHERE articleID = articles.id) / S.totalShares, 0)
              ) * 100 as trendingPoints
FROM        articles
CROSS JOIN  (SELECT sum(views) as totalViews FROM articles) A
CROSS JOIN  (SELECT count(*) as totalLikes FROM likes) L
CROSS JOIN  (SELECT count(*) as totalShares FROM shares) S
GROUP BY    articles.id
ORDER BY    trendingPoints DESC

这是最佳还是可以更有效地编写？

MySQL查询 - 如何根据＆＃34;视图＆＃34;，＆＃34;喜欢＆＃34;来获取趋势行。和＆＃34;股票＆＃34;从其他表？

更新

3 个答案: