This page contains both requirements and suggestions for your final portfolio for IM 1511. It will be finalized by the end of the week of 14 April. This version is complete. Any clarifications will be added in red in-line.
All work should be completed in your web hosting. If you choose to delete anything from your web hosting, please do not do so until after you receive your grade.
Your final portfolio should be organized in two separate pages in your WordPress. Each one might have subpages with parent-child attributes (see the bottom right when you are editing a page) to prevent them from being very long.
There are no explicit limits to the length of the two parts of the final portfolio. I would expect that it would be more than 1000 words (part one) and perhaps 750 (part two). Each part should contain 4 or 5 (or more) images that illustrate exactly the points you make in your writing. There are, of course, exceptions to this general rule.
Part One (30%)
Part One is the “doing” part of the final portfolio. It is an opportunity for you to take one of the methods we have experimented with during the semester and to expand significantly what you have already done, extend it to a new use case, collect new data for it or innovate it. It is your opportunity for you to show off how good you have become at experimenting with computers with humanities materials and to critique the methods that have been available to you.
Ideas for projects:
- Bring together a corpus of texts and classify them using the stylo package in R, studying the keywords of two distinct groups (ideas: women vs men writers; left vs right politicians; students of two nationalities or language groups).
- Choose an author and use Voyant Tools to generate a frequent set of words and build a simple Twitter bot that tweets snippets like that author (like the MAGA generator we experimented with).
- Build a corpus of texts of a particular genre (romance, science fiction, travel narratives, orientalist authors) and use Voyant Tools to examine it in depth. Combine your analysis with the oppose function, PCA analysis or stylo() to compare the results.
- Assemble a collection of images coming from one or two artistic movements (Impressionism, Cubism, Surrealism, etc) and automatically classify the images by color. Analyze the results.
- Assemble a collection of images (like our manga set) with repetitive features and use ARIES to manipulate and analyze the images.
- Make a more in-depth study of the English model on the Bushire Residency documents using our results with Transkribus.
- Choose a nineteenth century manuscript text from a digital library that resembles Bentham’s handwriting and run the public English model in Transkribus (or ours trained on the Bushire data) and evaluate the results.
- Expand your dataset from the data visualization exercise with ggplot to expand your understanding of the issues in the data. Use the documentation to explore what other things you can do with ggplot.
- Download your Facebook or Instagram photos from a expanse of time. Classify them using our scripts and examine the results.
- Choose an image dataset from Kaggle and classify them by color.
- Choose a non-English text of significant culture value. Use Transkribus to transcribe several dozen pages as well as the tags to mark up important parts of the textual structure. Discuss how well the platform allows you to work outside of the context of English.
- Visualize the French Enlightenment network data in R using the package of your choice. What are you able to learn about the period through your visualizations.
Some guiding questions (do not answer these one by one, but rather use them to think through your project).
- What part of the exercise is constituted of human labor and effort?
- What part of the exercise involves algorithms ?
- How did you have to prepare the data so that you could carry out the analysis?
- How does the combination of human effort and computational affordances extend your understanding of your data beyond what you alone would be able to see?
- How finished is your study? What kinds of challenges did you encounter? If you had more time how would extend the research you did?
- What patterns did you discover? Are they meaningful?
- What forms of bias can you ascertain in the processes?
- Go back to the site “What is Digital Humanities?” How has your understanding of the various quotes changed at the end of the semester and with your project?
Requirements in your writeup:
CODE: Please include any code that you employ in the body of your writeup, but do not just insert it into the visual text editor, as this has the tendency of distorting it. Choose a function that will display it well and not distort the code. There are a few plugins that will allow you to do this, such as Code Snippets. Using the text instead of the visual editor, you can also insert the tag <code> that would create a monospaced font as below. Explain how you have adapted the code to serve your own purposes.
hello <- "hello world"
hello
You can also try the tag <pre>.
hello <- "hello world" hello
DATA: Explain where the data comes from. Assemble the data in a folder and provide a link for download from Drive, web hosting or your own GitHub. You want someone to be able reproduce your experiment as best as possible
DISCUSSION: In your write up you should make sure that you discuss challenges that you faced, problems you had as well as any conclusions you were able to form. It should be a balance of product and process. If appropriate you links to connected you to former blog postings, readings, or any other web based materials. Use Zotero for any works cited.
VISUAL: Include any visuals by adding them as media in your WP.
Part Two (15%)
Part two is your “designing” or “reflecting” part of the final portfolio. It is your opportunity to think big about some valuable research that should be done in the world, without actually carrying it out.
The scenario: You have won a multimillion dollar grant. This means that you can assemble a multiyear, lab-based research group to investigate a question (or a set of interlocking questions) in a topic of interest to you in the digital humanities.
This part of the final portfolio may be connected to your Zotero bibliography, if you so choose. In any case, it should reference other projects of a similar nature. This is what is called in research an “environmental scan.”
Guiding questions (You do not need to answer all of these):
- What is the main goal of the project? To what theoretical issues in the digital humanities that you have encountered so far does it respond?
- With what theoretical issues outside the digital humanities does your project plan engage?
- How many people do you think would need to be involved? What kinds of expert knowledge would you include? What kind of crowd knowledge would you seek to tap into?
- How social is the project? Does it engage the crowd?
- How algorithmic is the project? How much does it rely upon machines?
- With what kinds of data would the project work? How were they acquired? Are they open data? Are there any ethical or legal considerations in acquiring the data?
- What kinds of analysis of the data would you like to do? How do you imagine you would disseminate that data?
- What visuals that you could make would help you illustrate the project?
- If you were making a website to explain the project, what would the landing page of that site look like?
Requirements:
There are very few specific requirements for this part of the project. Be as creative as you can. Balance textual description and visuals.
For inspiration, you might look to a few places where innovative new work is taking place:
The program for the DH 2019 conference in Utrecht, NL : https://www.conftool.pro/dh2019/sessions.php
Pages 61- of D_H (Burdick et al). This section is called “A Portfolio of Case Studies.”
VERY LAST STEPS: Create a backup of your web hosting.
Instructions here.
Nothing will be archived without your permission–if you choose to keep your work private, it will not affect your grade. I would still like to have a copy of it.