cogs108/projects

Final Project materials and description.

Projects

Materials for the Project in COGS108.

Project Documentation

• Project Guidelines
• Team Policies

Project Templates

Templates have been provided in your group's project repo.

• Proposal: ProjectProposal_groupXXX-Fa22.ipynb
• Checkpoint #1: DataCheckpoint_groupXXX-Fa22.ipynb
• Checkpoint #2: EDACheckpoint_groupXXX-Fa22.ipynb
• Final Report: FinalProject_groupXXX-Fa22.ipynb

Final Project Checklist

Students often ask for a rubric. You can use this checklist to help guide your thinking on the final project. If you check off all the boxes below, you should be in good shape to get a perfect score on your final project.

Overview, Question & Background

Overview:

• Write a clear summary of what you did
• Briefly describe the results of your project
• Limit overview to 3-4 sentences

Research Question:

• Include a specific, clear data science question
• Make sure what you're measuring (variables) to answer the question is clear

Background & Prior Work:

• Include a general introduction to your topic
• Include explanation of what work has been done previously
• Include citations or links to previous work

Hypothesis:

• Include your team's hypothesis
• Ensure that this hypothesis is clear to readers
• Explain why you think this will be the outcome (what was your thinking?)

Dataset(s):

• Include an explanation of dataset(s) used (i.e. features/variables included, number of observations, information in dataset)
• Source included (if outside dataset(s) being used)

Data Analysis:

Data Cleaning & Pre-processing

• Perform Data Cleaning and explain steps taken OR include an explanation as to why data cleaning was unnecessary (how did you determine your dataset was ready to go?)
• Dataset actually clean and usable after data wrangling steps carried out

Data Visualization:

• Include at least three visualizations
• Clearly label all axes on plots
• Type of all plots appropriate given data displayed
• Interpretation of each visualization included in the text

Data Analysis & Results:

• EDA carried out with explanations of what was done and interpretations of output included
• Appropriate analysis performed
• Output of analysis interpreted and interpretation included in notebook

Privacy/Ethics Considerations:

• Thoughtful discussion of ethical concerns included
• Ethical concerns consider the whole data science process (question asked, data collected, data being used, the bias in data, analysis, post-analysis, etc.)
• How your group handled bias/ethical concerns clearly described

Conclusion & Discussion:

• Clear conclusion (answer to the question being asked) and discussion of results
• Limitations of analysis discussed
• Does not ramble on beyond providing necessary information

Video:

• Question asked is clear to listeners
• Effective visualizations presented
• Clear explanations throughout
• Take home message clear
• Within 3-5 min time limit

Final Checks:

• Edit all text for clarity
• Remove all instructions
• Be sure text included throughout to guide reader
• Check to make sure all text and images are visible
• Names included
• Renamed file : FinalProject_groupXXX-Fa22.ipynb, where 'XXX' is replaced by your group's group number

After the course is done:

• If you checked YES to make project public: the final project notebook (and only that!) will be placed in a repo with the rest of this quarters public reports. This helps future students by providing examples!
• Your projct repo will remain available to you in the near future. We cannot guarentee that will always be the case. That repo will never be public.
• If you would like your own copy of the entire repo you should follow these instructions: https://docs.github.com/en/repositories/creating-and-managing-repositories/duplicating-a-repository Once you have done that it is yours forever. You will also be able to control access to the mirror (make it public or private as you would prefer)

---

License

The content of this project itself is licensed under the Creative Commons Attribution 3.0 Unported license.

projects's Issues

Conflict in README

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

Test issue

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

Test issue

test

Guidelines: Broken Links

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 :)

Test issue

"{"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 | "}}"

Test issue

	group	grader	Question	Background	Hypothesis	Data	Ethics/Privacy	Expectations & Timeline	Overall	Total
0	0	example	15	15	10	25	20	5	10	100
1	1	Atman	12	7	9	15	15	5	8	71
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?

Project.c

hi

Ignore me

Project Organization

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.

Test issue

	group	grader	Question	Background	Hypothesis	Data	Ethics/Privacy	Expectations & Timeline	Overall	Total
0	0	points possible	1.2	1.2	0.8	2	1.6	0.4	0.8	8
1	1	Atman	0.96	0.56	0.72	1.2	1.2	0.4	0.64	5.68

====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?

Test issue

	group	grader	Question	Background	Hypothesis	Data	Ethics/Privacy	Expectations & Timeline	Overall	Total	Comments
0	0	example	15	15	10	25	20	5	10	100	Question
											Background
											Hypothesis
											Data
											Data Ethics/Privacy
											Expectations & Timeline
											Overall

Test issue

	group	grader	Question	Background	Hypothesis	Data	Ethics/Privacy	Expectations & Timeline	Overall	Total
0	0	example	15	15	10	25	20	5	10	100
1	1	Atman	12	7	9	15	15	5	8	71
---:	:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
1	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?