Problem of data visualization (Dianping.com)

hupili / python-for-data-and-media-communication-gitbook

An open source book on Python tailed for communication students with zero background

115 stars 62 forks source link

Problem of data visualization (Dianping.com) #92

Open CelineTsang opened 5 years ago

CelineTsang commented 5 years ago

Problem of data visualization (Dianping.com)

Describe your environment

Operating system:
Python version:
Hardware:
Internet access:
Jupyter notebook or not? [Y/N]: y
Which chapter of book?:

I grabbed the restaurant information of dianping.com/Hong Kong (including name, address phone number, tags, feature, average price), but since each data is unique, such as the price does not exist range, are specific Numbers. tags: e.g.标签：粤菜中餐茶餐厅米其林三星 This makes it difficult to visualize because there is too much text to display complete information and so on. I wonder if you have any ideas and solutions to offer?

The original csv is like this:

hupili commented 5 years ago

Please leave your notebook link and better the NBViewer link since there are dynamic charts.

hupili commented 5 years ago

For your dataset, followings are suggested:

transform the price column from str to int. Once done, 1D analysis on this column can already yield to multiple story points, e.g. distribution/ min/ max/ ...
For the tags column, you may want to convert it to numeric encoding. One way is to test keyword for each cell and put 1 if the keyword exists, or 0 otherwise. The process is very similar to the first image in your question body. Another reference is this top name in tweets