Project Link: Surveilling Data Weapons

I gained experience with text mining in a Corpus Linguistics course; we primarily used AntConc in that class. In my final project, I created my corpus and used AntConc to analyze the context in which journalism/media outlets write about digital activism. The corpus totals 108 articles and 148821 tokens. There are 18 articles from The New York Times, BBC, 2 from The Washington Post, 4 from The Wall Street Journal, 14 from CNN, 1 from the Financial Times, 18 from Al Jazeera, 4 from PBS, 1 from Foreign Policy, 5 from NPR, 37 from The Guardian, 2 from Reuters, and 2 from CNBC. For the in-class exercise, I used Voyant to analyze the same corpus. 

With Voyant, I enjoyed the summary the software provided of the documents; document length, vocab density, avg. words per sentence. Additionally, Voyant has more visual features, making the scale of the words more understandable (whether there is a larger quantity of words or phrase structure that has been repeated). 

Connecting GitHub was convenient and something that AntConc could not do; however, I couldn’t connect or compare my corpus to a reference corpus (which AntConc can do). I could compare the frequency of words found in the texts to one another (ex: “digital activism” vs. “internet”). For basic text mining for someone with less knowledge, Voyant is easy, convenient, and powerful. 

The corpus summary.

The word cloud created by Voyant.

The collocates of the term “digital activism”.

The relative frequency of “internet” vs “digital activism” within the documents of the corpus.

A small workshop held by GCDI, From Paper to Website, was informative and helpful – especially with the upcoming final project. Taught by Silvia Rivera Alfaro, the workshop was inspired by her own experience creating a website (Ni Unos Mas: A Collective Identity in Rebeca Lane’s Rap) for one of her graduate papers. The workshop discussed translating a paper into a website, making it both digestible and accessible, regardless of the audience. Throughout, Alfaro wanted us participants to remember three core questions:

1. What is text and what are the different types of texts?
2. What choices do I need to consider to effectively translate an academic paper into a website?
3. What could be the benefits of a research paper being translated into a website?

In the first half of the workshop, we discussed the structure of an academic paper, and how to translate that structure into a website. This included considering the audience, the tools we are providing, how we want that tool to be used, the interactivity of the page, etc. In the second half, it became more hands-on. Because there were only myself and two other participants, we got to work individually – we analyzed an academic article and tried to visualize the article as a website. During the last 10 minutes, we shared our work and explained our thought process. 

Ultimately, the workshop was tight-knit and fun. I’m happy to have gone because it’s making me consider more conditions for my DH project and how I want to share that with the world.

My research interests lie in digital activism and social movements; activists have increasingly turned to digital platforms to mobilize the public around various causes. Using X (formerly Twitter), Instagram, and more recently, TikTok, many viewers become more aware of social movements through their screens. I am currently conducting an independent research project, The Education of Digital Activists (IRB approved) that inquires about how digital activists in college (or those who have graduated within the last 5 years) who have been exposed to critical social theory understand their own privileged, their relationship to social issues, and it impacts on their digital activism. Through my interviews, I’ve noticed that all my interviewees have been activists of color. However, through broader and general social media use, the faces of many digital activists’ movements/accounts feel white. 

A working title, Mapping Digital Activism Through Blackness and Solidarity focuses on mapping digital activist movements of color, specifically those belonging to black digital activists, and the movements they support. For reference, Black-Palestinian solidarity has been a relationship that has existed for decades, referencing the relationship between black people within and outside of America and Palestinians have had with one another through a belief in the “joint struggle”. My project aims to bring visibility to different digital activist movements of color globally, possibly showing their activity on a map.

I’m thinking of creating a website; maps would be displayed using OpenStreetMap. University of Minnesota’s project ‘Mapping Prejudice’ (https://mappingprejudice.umn.edu/) includes a sort of animated time-lapse feature that I enjoy. I intend to have a similar sort of animation feature within my own project. For this, I anticipate that QGIS may be used to create a shapefile(s) containing a multitude of points for each activist movement chosen. Moreover, I have heard of activist datasets that may fit my needs and so I will search for them. However, my overall goal is to create the datasets myself. 

I’m open to exploring other options for the mapping, as well as any available datasets or avenues to narrow down my interests here.

Data Sources: STRI GIS Portal

Last week, I introduced the fact that my mum is British-Guyanese. On top of that, my father is Panamanian. He comes from Bocas del Toro, an area where Chiquita bananas (owned by the United Fruit Company) were produced. In January, we visited his hometown together for the first time; he was surprised at the changes. Entering the port via water taxi, we passed massive ships carrying cargo. Cranes could be seen on the land lifting containers up and down. My father could barely recognize where he was; he remarked that Chiquita must have paved the roads to make transportation easier for themselves.

I thought of Panama not just because of my experience there earlier in the year, but because of the climate issues that it is facing. Gardi Sugdub, an island off the coast of Panama, is sinking (due to rising sea levels). Primarily a fishing town, it’s home to an indigenous community. This past summer, Panamanian officials relocated locals.

Initially, I wanted to create maps tracking the changes in climate and sea level over Gardi Sugdub and surrounding islands. Unable to find the specific files that I wanted to use, however, I decided to begin by making basic maps of the general country within QGIS.

QGIS is a free, open-source software used for mapping and geographical information systems (GIS). Without some sort of tutorial or how-to, it can be confusing. Because I’ve used QGIS before to map NYC census data, I treated my map-making as a “refresher”. However, even with the previous experience, it is not as “common sense” as it seems. My first time using it, I didn’t understand the meaning of a “Vector Layer”. Additionally, I had to google “how to make a map QGIS” to get from the “project” mode to the “layout” mode. After using the software a couple of times for basic map-making, the process becomes almost automatic.

I used data from the Smithsonian Tropical Research Institute: GIS Data Portal. Because the chosen data is so organized, I didn’t have to do any JSON edits or cleanup. I made four distinctive maps: Provinces of Panama (2016), Areas in Panama With Electricity (2010), Areas in Panama Without Electricity (2010 – Modified Legend), and Areas in Panama Without Electricity (2010 – Original Legend).

Provinces of Panama (2016) showcased the general administrative areas set by the government of the provinces within the country. I have previously never seen Panama’s indigenous provinces or areas mentioned on a map or in general conversation, and so when turning on the “labels” of the different areas, I was happily surprised. However, with the labels, I found it difficult to position them exactly the way I wanted; it was either they hovered over the areas and obstructed outlines, or they overlapped with one another – making it confusing to read.

With population information from the 2010 census, Areas in Panama With Electricity (2010) displayed the general areas in Panama where there was electricity. STRI did not specify whether it was people, households, or entire neighborhoods that had electricity; the distinction is important as the count would drastically change. Moreover, it was on this map that I was not entirely sure whether it was my skills, the information data, or the spread of the population that was leading the map to look the way it did. The two areas where electricity seems at its highest were near the capital, Panama City, and Chiriquí, where many power plants are located. To modify the legend, I changed the styling to my “population” layer from a single symbol to graduated, with an equal interval for my VPOCLUEL (with electricity) value.

I treated Areas in Panama Without Electricity (2010 – Modified Legend) and Areas in Panama Without Electricity (2010 – Original Legend) as experiments, due to the different highlighted areas. After changing the legend values, there were no areas on the map displaying any sort of color values. This would suggest that all areas seemingly had electricity. However, in the original legend, the values for those without electricity were low, with the last spread ranging from 317 – 397. Having this sort of legend spread didn’t make sense considering the last legend contained values in the thousands.

Ultimately, I want to continue exploring QGIS features, such as the plugins, the Python and SQL servers, etc. Ujaval Gandhi makes great step-by-step tutorials ranging from basic to advanced and with time, I’ll be following those.

Google Slides: https://docs.google.com/presentation/d/1oJx6LofrL0rzZkqGGs_qOxHh42Nr073Mqq2xcvScrAM/edit?usp=sharing