我如何简单地将数据分成组，以便我可以对它们进行操作

Question

因此，我经常遇到必须对数据进行分区或分组（逻辑上）然后进行比较的情况。比如：

“某个值出现在组中和不在组中的频率以及它们各自的百分比是多少”

select LESS_THAN, GREATER_THAN, round(LESS_THAN /(LESS_THAN+GREATER_THAN),4) as percentage
from (
    select 'less than' as type, level as rn from dual where level <= 5  connect by level <= 10 union
    select 'greater than' as type, level as rn from dual where level > 5 connect by level <= 10
)
pivot(
    sum(rn)
    for type in (
        'less than' as LESS_THAN, 'greater than' as GREATER_THAN
    )
);

有什么聪明的方法可以简化这类事情吗？

Answer 1

您可以使用条件聚合：

SELECT COUNT(CASE type WHEN 'less than' THEN 1 END) AS less_than,
       ROUND(AVG(CASE type WHEN 'less than' THEN 1 ELSE 0 END), 4) AS percentage
FROM   table_name

对于样本数据：

CREATE TABLE table_name (type, rn) AS
  SELECT CASE
         WHEN level <= 5
         THEN 'less than'
         ELSE 'greater than'
         END,
         level
  FROM   DUAL
  CONNECT BY LEVEL <= 10;

输出：

少于	百分比
5	.5

小提琴