【数据分析师_02_SQL+MySQL】999_阶段测试（题目及参考答案）

---

转载请注明出处 ！
手动反爬：阶段测试（题目及参考答案）
https://blog.csdn.net/Lyun911/article/details/111829621

注：以下使用 MySQL 演示

---

# Author：DA-7-163 Lyun
# Date：2020-12-27
# Notes：去除图片后，以下内容可在 Query中直接运行

# 1.1 创建db名字叫 flight 创建新表名字叫on_time_performance
# 1.2 航班数据检查csv ，导入到sql数据库中

# 1.3 执行count语句查看总行数
SELECT count(*) AS row_num FROM on_time_performance;

# 2.1 在航班及时率表中有如下delay延迟字段，请找出导致延迟次数最多的是哪一个原因 ，并给出各个原因累计的延迟时间 WeatherDelay, CarrierDelay, NASDelay, SecurityDelay, LateAircraftDelay
-- DROP VIEW Delay_VIEW;

CREATE VIEW Delay_VIEW AS
SELECT 'WeatherDelay' AS REASON, COUNT(WeatherDelay) AS DELAY_TIMES, SUM(WeatherDelay) AS DELAY_HOURS FROM on_time_performance WHERE WeatherDelay > 0
UNION
SELECT 'CarrierDelay', COUNT(CarrierDelay), SUM(CarrierDelay) FROM on_time_performance WHERE CarrierDelay > 0
UNION
SELECT 'NASDelay', COUNT(NASDelay), SUM(NASDelay) FROM on_time_performance WHERE NASDelay > 0
UNION
SELECT 'SecurityDelay', COUNT(SecurityDelay), SUM(SecurityDelay) FROM on_time_performance WHERE SecurityDelay > 0
UNION
SELECT 'LateAircraftDelay', COUNT(LateAircraftDelay), SUM(LateAircraftDelay) FROM on_time_performance WHERE LateAircraftDelay > 0
ORDER BY DELAY_TIMES DESC;

SELECT * FROM Delay_VIEW;
# 结果显示：延迟最多的原因是 CarrierDelay，共计46,191次，共计1,639,773小时

# 2.2 延迟时间倒数第二的是哪个原因 ，给出累计的延迟时间
SELECT * FROM Delay_VIEW 
ORDER BY DELAY_TIMES
LIMIT 1, 1;
# 延迟时间倒数第二的是WeatherDelay，共计5,140次，共计249,979小时

# 3.1 请显示2015-01-03日的航班数据，要求显示AirlineID, FlightNum, Origin, Dest, DepTime, TailNum, AirTime, Distance如上字段
# 3.2 由于时间部分查询非常多，请在时间字段上增加索引，加快查询速度，此处请截取两张图片，比对执行速度的区别
EXPLAIN
SELECT FlightDate, AirlineID, FlightNum, Origin, Dest, DepTime, TailNum, AirTime,Distance
FROM on_time_performance WHERE FlightDate = '2015/1/3';

# 3.3 由于数据很多，写出快速获取到第50-55行AirlineID, FlightNum, Origin, Dest, DepTime, TailNum, AirTime, Distance数据所用的sql语句。同时给出第50-55行 截图；
SELECT FlightDate, AirlineID, FlightNum, Origin, Dest, DepTime, TailNum, AirTime,Distance
FROM on_time_performance LIMIT 49, 6;

# 4.1 探究一年内飞行航班的季节性，并统计出12个月的航班总数
SELECT Quarter, COUNT(*) AS Count_Q 
FROM on_time_performance 
GROUP BY Quarter 
WITH ROLLUP;
# 结果显示：第3季度为旺季，第1季度为淡季。
# 12个月的航班总数为469,008

# 4.2 找出航班最多的一个月是几月份
SELECT on_time_performance.Month, COUNT(*) AS Count_M 
FROM on_time_performance 
GROUP BY on_time_performance.Month 
ORDER BY Count_M DESC
LIMIT 1;
# 结果显示：航班最多的是7月，航班数为41,852

# 5 现有一个json文件的airplane数据，写一段python程序把数据转换成csv，导入数据库中。然后统计飞机最多的两个生产商

# 结果显示，生产飞机最多的是BOEING和AIRBUG