Figure 1
The number of tokens per year in Courier (non-curated). A token is a compound of characters separated by spaces.
Figure 2
Courier curation workflow: This maps the processes applied to create the Curated Courier 1.0 package.
Figure 3
Print screen image of the Notebook feature “Load Topic Model”, showing the 11 topics that contain the word “society” among their 20 most likely words from a model of 200 topics.
Figure 4
Print screen image of the Notebook feature “Topic-Word-Distribution”, focusing on topic 143 (social systems) with its 30 most likely words, each assigned a probability value.
Figure 5
Print screen image of the Notebook feature “Topic Trends over Time”, showing the trend line of topic 143 (social systems), using an average weight measure.
Figure 6
Print screen image of the Notebook feature “Browse Topic Documents”, focusing on topic 143 (social systems). Choosing weight threshold 1.0, time period 1975–1990, and only articles that explicitly have “UNESCO” in the author tag, generated six such articles. The article “The General History of Africa” (1979) by former UNESCO Director-General Amadou-Mahtar M’Bow is highlighted, with part of the article text shown to the right, including a link (“PDF”) to the corresponding Courier issue.
Figure 7
Print screen image of the Notebook feature “Topic Topic Network”. Different options are possible for filtering the structure and layout of the topic network. Node labels represent topic numbers.