目的:爬取客路旅行公开的数据,通过数据分析《台北101观景台门票》活动的现状、风险与预测销售量
网址:https://www.klook.com/zh-CN/
目标:台北101观景台门票
(https://www.klook.com/zh-CN/activity/1659-taipei-101-taipei/?krt=r20&krid=63cf4981-29eb-47fa-5413-cde7fb8c7a7e)
人群:大陆游客
时效:2019-02-14
工具:Python、Mysql、Excel
一、风险
①、风险用户
select author,count(author) from infos where title='台北101观景台门票'and rating <=60 group by author order by count(author) desc limit 10;
通过Mysql查询:评分<=60,排行前10用户
select author,count(author) from infos where title='台北101观景台门票' and author in ('Karen','Ka Man','Yiwen','Chung Chih','Lai Yee','Yu chen','Wing Fung','Sze Nga','Siu Man','Sook Yin') group by author order by count(author) desc limit 10;
通过Mysql查询差评用户参加活动的次数:差评前10的总参与活动数
分析差评用户的差评率,红色用户存在较大优化空间或存在较大风险
②、差评词云图
差评主因:天气|排队二、机会
①、优质用户
select author,count(author) from infos where title='台北101观景台门票'and rating >=80 group by author order by count(author) desc limit 10;
分析好评前10用户:该用户参与活动次数高,是活跃或重点用户
select author,count(author) from infos where title='台北101观景台门票' and author in ('Michelle','Jennifer','Michael','Karen','Mary Grace','HENG YANG','Richard','David','PENG','Ryan') group by author order by count(author) desc limit 10;
分析好评用户的好评率,红色用户存在宝贵的反馈建议
select author,count(author) from infos where title='台北101观景台门票' group by author order by count(author) desc limit 10;
通过Mysql查询:参与活动次数最多的用户是重点用户
②好评词云图
好评主因:方便|便宜
三、现状
好评数:6154+1723=7877 好评率:92.4%
网友评论