James Clawson

@jmclawson.bsky.social

58 Followers 74 Following

Once academic (Digital humanities). Now in Chicago. #DH, #Data, #Dogs, #RStats, #Running — hashtags’ songs: Cicadas on blue. jmclawson.net github.com/jmclawson

Code and output preparing a table of word counts for documents in a corpus, with document titles depicted in italics.

Code:
corpus_joyce |>
tabulize()
italicize_titles()

Output shows a table with three rows and two columns, with the first column for title and the second for word count. Dubliners is shown with 67,945 words, A Portrait of the Artist as a Young Man is shown with 84,926 words, and Ulysses is shown as having 264,975 words.

R code and output, showing the creation of a methods narrative from a log that was created behind the scenes.

R code:
bigrams_joyce |>
narrativize(format = "text", person = "we", return = "html")

Output:
First we retrieved a corpus from Project Gutenberg using ID numbers 2814, 4217, and 4300. Then we moved column `subsection` into body text. Next, we loaded texts by tokenizing words, converting to lowercase, and preserving paragraph breaks. We identified documents using the column `title`. Finally, we constructed bigram sequences from the text.

Map focused on Lakeview neighborhood in Chicago. Three red circles show the locations of Red Line trains heading to Belmont station.

Small display beside a speaker. On the display are colored rows of purple, red, and brown, with destinations and estimated times of arrival into Belmont station—listed at the bottom beside a toggle with an arrow pointing down.

Scatter plot showing three clusters: travel writing at the far right, commercial fiction at the bottom left, and literary fiction at the top center. A few texts appear as outliers, positioned closer to other clusters.

Screenshot of Shiny app comparing structure and performance of one data set using Parquet files, duckDB, SQLite, and MySQL. The “Structure” page is selected, and additional pages labeled “Query” and “Languages” are also listed

Bar charts comparing NRC sentiment distributions for Bluesky posts discussing Python and RStats. In Python posts, the sentiments fear, trust, and anticipation show the highest scores of about 1.25%. Among RStats posts, the sentiments trust, anticipation, and joy score the highest, at about 2%, 1.7%, and 1.1%.