cogs108 / projects Goto Github PK

Working collaboratively in notebooks is not fun. May be worth thinking about a better structure that would lead to fewer version control issues.

Notes:

• Issues may not actually be an issue much longer: https://twitter.com/DynamicWebPaige/status/1397224065077850115
• Adding additional structure, i.e. Cookie Cutter Data Science may be worth building into the course

Don't have a real solution yet, so nothing this here for now.

Question |
Background |
Hypothesis |
Data |
Data Ethics/Privacy |
Expectations & Timeline |
Overall |

"{"group":{"0":0},"grader":{"0":"example"},"Question":{"0":15.0},"Background":{"0":15.0},"Hypothesis":{"0":10.0},"Data":{"0":25.0},"Ethics\/Privacy":{"0":20.0},"Expectations & Timeline":{"0":5.0},"Overall":{"0":10.0},"Total":{"0":100.0},"Comments":{"0":"Question | \nBackground | \nHypothesis | \nData | \nData Ethics\/Privacy | \nExpectations & Timeline | \nOverall | "}}"

hi

test

Just a note - some of the links in the Group project guidelines page are broken (I think just some links to other files within the Group folder).

I would / could open a PR to fix, but I'm not sure if there some things might move sometime soon anyways, so for the moment, just a note to check and fix this at somepoint :)

There's a conflict in the readme... it looks like someone added the word "an" here: https://github.com/COGS108/Projects/blob/master/README.md#data-analysis

====Question | Unclear from the question or the background how "monotonous" will be measured. It's clear that you'll be analyzing lyrics and determining themes from those lyrics. And, it's clear how one may approach measuring the "repetitive" nature from the songs. But, unclear exactly how monotonous will be measured
Background | This section not fully thought out nor explained. What evidence do you have to suggest that the genre is monotonous? What other research has been done on this before (Note: encouraging you to check out pudding.cool and work they have done on this topic if you have not already) This background work should be introduced, summarized, and linked to directly. What have others written about to this "instant gratification" and "shallow" nature? This shoould all be explained here
Hypothesis | "Gratification" trend not described and remains unclear. And, this hypothesis only loosely relates to the question. Themes not discussed in hypothesis but a main component of your question, for example.
Data | Genre keywords you' would want not specified. Also, is this all released songs within genre? All songs to hit a specific chart? What would ou want? Why 15,000 songs? Which 15,000 songs?
Data Ethics/Privacy | How will you check validity of your song lyrics? What if certain songs aren't in your databased? How does the "readibly available" nature of genius limit your analysis? What will you do to ensure your analysis is valid? What checks on your data and results would you need to do to ensure the accuracy and validity?