Skip to content

Latest commit

 

History

History
56 lines (45 loc) · 4.58 KB

guestbook_patrick.md

File metadata and controls

56 lines (45 loc) · 4.58 KB
Raw

Analysis of Russian News Sites

Patrick Stoyer

Katie's post

  • What was done well? Very good explanations and walking through the code. A lot of it is stuff I've never done before, so I like how it's relatively easy to follow along.
  • What can be improved? When you get to the point of annotating for sentiment (and you might already be planning this), I'd suggest printing out translated snippets too, just so that the reader can understand the different qualifications fo r the different types of sentiment.
  • One thing I learned: I learned how to use BeautifulSoup to get text from a webpage, and how complex this can actually be! I'm sure there's a lot of room for small errors, so I appreciate the fact that you made a separate script for each file.

Matt

  • One thing I liked : Decent explanations of code, presents errors in their full glory for all to see. In other words, actually goes over what's wrong instead of trying to hide it.
  • Area of Improvement : Presentation - you print out your scraping code, rather than copying and pasting it into a cell (be it code or markdown with syntax highlighting) or uploading the file separately? I fail to see what this accomplishes, if anything makes things worse. Code with plain text fonts just looks ugly and hard to read. I would at the least put it in a markdown cell with syntax highlighting.
  • One thing I learned : BeautifulSoup and webscraping. Also, using os change directory in python to navigate to files in a live program

Eva's post

  • What was done well: Good organization and pretty good explanations throughout your jupyter notebooks and in your project plan. Good job with the web scraping, you got a lot of data! There's a lot of different places to go with your data and I'm interested to see what else you discover through topic modeling and sentiment analysis especially.
  • Improvements and suggestions: It's unclear where to start when looking through your project plan and progress report. It could be helpful to include links throughout your write-up documents to different jupyter notebooks. You could also put a short explanation in your readme of what's in each folder, document, etc. Explanations (mostly just for the LDA/LSA) could have been more detailed.
  • What I learned: It's a really good idea to lemmatize the text before topic modeling! I didn't even consider that I could do that. I'm going to try that in my own analysis now.

John (accidentally did the same two twice for to-do 1 and 2)

  • What was good: You walked us through everything really well (though maybe you don't need to flash the whole text as much)! I loved how you showed us how your code worked and everything in your overviews.
  • What could be improved: I would like some interpretations of your data as you go along! I think you're touching on some great things, but some things are still lacking in depth.
  • What I learned: Data scraping is really interesting. The code seems to have worked well for you, and I'm interested in using selenium in the future!

Cassie's entry:

  • What I liked: Your code is well-organized and annotated with good explanations of what you're doing. Also, your approach to data acquisition was super thorough! Can't wait to see what info you get from your analysis!
  • What Can Be Improved: Your project reports could use a few more links to make your repo more navigable, and some more description on what you're analyzing (like perplexity and coherence).
  • What I Learned: A heck of a lot about webscraping! You really use advantage of the trees/hierarchies. I've only just touched the surface of web scraping, so your code was really informative!

Tingwei's entry:

  • What I liked: Your Jupyter notebook is really clear and easy to follow. The repository is well-designed. It looks like a professional project task.

  • What Can Be Improved: It will be better to add more instructions on how to read the results you provide.

  • What I Learned: I learned how to organize the project, and process multiple data together. This may be closer to real work situation.

Elena's entry:

2019.04.11

  • What I liked: You pulled in a lot of data to work with! That's awesome, especially considering you're modeling for your project.
  • What Can Be Improved: I was a little confused about the difference between coherence and perplexity score in your last jupyter notebook file. Maybe add a little more information about that?
  • What I Learned: How to use gensim, and how to effectively webscrape to get a lot of data.