Your goal with this project is to explore a specific aspect of algorithms and the arts that you want to learn more about and share what you learn with a specific audience. The structure of this assignment is designed to help you find answers to your questions by finding and reading reliable resources that already exist (TONS has been written on this topic) AND conducting close analysis of a specific recommendation engine on your own.

Process

Your first step will be to prepare notes as you explore (research). The prompts for note-taking below are offered as a guide for this work, but I ultimately want you to find a format for note taking that works best for you.

Step 1: Explore our Zotero library for reliable resources related to your interests

The first place to look for this project is a library of almost 200 resources (articles, videos, interactive tutorials) related to algorithms and the arts. I have assigned some of these resources for all of us to read together this semester, but there are many more that are fantastic and may prove useful as you get started with your project. To access the library, visit this link. You can see a list of tags I have created for the resources in the lower-left corner of the library. This tagging system is a work in progress. One possibility for this project is that you develop additional tagging structure for this library (How is a library like a streaming platform? How is an educational resource like a work of art?)

Step 2: Find and evaluate resources that can help you answer your specific questions

We will discuss research strategies in class and you will share resources during our discussions, but your goal at this stage is to find the best sources out there. To ensure you’re pushing through the easy-to-find-but-not-very-reliable things out there, I am asking you all to take notes as you research. Please include answers to the following questions about each resource you find to determine if the information it conveys is reliable. When your research notes are due, please upload a file addressing these questions for all of the sources you have found by that date.

• Full title of the source and the larger work it is a part of (a single web-page on a website, a story in a book, a song on an album, etc.). It’s best to write the title in accordance with a citation style like MLA or Chicago.
• Describe the genre of this source. What is it (blog post, peer-reviewed article, YouTube video, magazine article, textbook) and how does that matter for its reliability?
• Describe the author/creator, starting with the full name. If it was created collaboratively, try to include all of the people who contributed in a significant way to the creation of the source. For each significant contributor, include noteworthy bits of biography. Then explain if/how these things matter in determining if the resource is reliable.
• Describe the publisher, starting with the name of the publisher. Is this a well-respected source of news? An academic press? An individual who has self-published on the Internet? A company more interested in selling a product other than the resource? What does the publisher tell you about the quality of the information presented in the source? Does the publisher make you think this source is appropriate for your project?
• What do you think the purpose of this source was when it was first published?
• How are you thinking of using this source for your project? It will be most helpful if you include particular page numbers or timestamps for material you might cite in your project. You are also welcome to include a link to another document you’re using to take notes (Zotero is one place to do this).

Step 3: Analyze a specific recommendation engine

We have looked together at many streaming platforms and recommendation engines (Netflix, Goodreads, Storify, Pandora, Spotify, YouTube Music, Bandcamp, and Pinterest). Because most of these platforms are proprietary, we are not fully able to see exactly how their data is organized or how algorithms are designed to shape user experiences. He have nevertheless tried, with the help of some examples, to identify the organization of data and test theories about how these systems are structured. One component of your final project should involve close analysis of a recommendation engine of your choice. This might mean looking more closely at a recommendation system we’ve already discussed or exploring a new one we didn’t discuss. The following questions can guide your analysis. NOTE: you’ll need to find reliable sources to answer many of these questions!

• What is the purpose of this recommendation engine?
• What features does the algorithm use to sort creative works?
• Does it use machine learning?
• If yes, what data was used to train the model(s) used?
• What unintended consequences have resulted (or could result) from widespread use of this recommendation engine?
• Has anyone tried to exploit or “game” this engine?
• What else is interesting to you about this recommendation engine?

Step 4: Present your findings

You will be creating a post on our course website to share your final project, but you can create that post in many different ways.

• Compose a blog post sharing what you have learned with future students
• Prepare a new unit for this course (on recommendation engines for games?) complete with suggested readings, a shared creative work for discussion, and a template for building a spreadsheet
• Create a set of labels and controlled vocabulary for a particular grouping of creative works (an easily missed subgenre that you wish more people could encounter?) and embed it in your post with an explanation of how you’ve organized it and how you hope it would fit into an existing or new recommendation engine.
• Revise three or more reviews you’ve already written and share them, with spreadsheets and commentary on how recommendation engines could better promote the creative works you’ve reviewed
• Prepare an annotated bibliography (or, list of amazing resources with your commentary) about your area of interest so future students can explore what you’ve found and then take the research further (this can be something you contribute to our existing Zotero library or something you create as a separate system)
• Something I haven’t thought of? Let me know what you’re thinking!

Your final project should be published in WordPress, but you can use any external tools you would like to produce content that can be embedded in that post (canva, Photoshop, YouTube, etc.). When you’ve completed revisions to your post, please upload to Canvas a brief project description that addresses your decisions about:

• Purpose: State the purpose of your project explicitly, even if you didn’t state it in this way in your post.
• Citation and Attribution: Describe how you approached citation and attribution and how you think your decisions worked.
• Audience: Include a note describing the audience you are hoping to reach, the decisions you made to best reach them, and the steps you plan to take to help this audience find your work.

Note: You are not required to make your post visible to classmates or the public. You also have the option to change the way your name displays if you want to make your work public on our site but not directly linked to you (it will still be obvious to viewers that the author was a person enrolled in my course in a particular semester). Please send me an e-mail if you want guidance on this. Also note that you can refer to a guide to finding, reproducing, and adapting openly licensed work as you compose your blog post. This is an important step if your post will reproduce images or remix existing resources.