I crawled some data before and did not analyze it. Today I am practicing the use of jupyter notebook. I simply took the 7-day hot data crawled on February 7 for simple analysis, mainly using mongodb's pipeline pipeline and other technologies plus charts Use of the package.
The first time I looked at the effect, I felt that it was the title party (a good 7-day trend). It may be because some articles are so good that they have been kept, and then articles older than 7 days are removed.
The number of inclusions is average every day, and the reason for the low on February 6 may be that it is too late to count (I am a crawler on February 7). Look at the specific time of the release:
Most articles are written in the afternoon and evening. Most of them are for work reasons. (In fact, I really admire those who go to work during the day and analyze and study for everyone at night).
Lively look at the author of the article:
By analogy, statistical data such as the most rewards can also be analyzed. Finally, take a look at the article with the highest overall ranking (comment, browse, tip and like):
last blow! In the 2016 Mac series, I think I think the most recommended software learner's notes, the failure of education? In the past few days of the Spring Festival, 7 movies I watched once told me good night and Spring Festival gala sketches. It’s not funny. I can bear it. People are wrong. Jane 14: Is secret love really a matter of one person? Recommended non-utilitarian English learning tools (iOS version) Three-minute impromptu speech and monthly income of 10W+. These are the skills you have to learn. How did I counterattack from Tufeiyuan to Baifumei? Not happy to marry a poor man? I am the woman sitting on the bike laughing
It seems that everyone likes to read some soothing articles (I'm already crying in the toilet).