I hope you are well. I had a few questions about the meanings of certain terms and diagrams because I could not find the answers on the documentation pages.
- After running the “Terms Extraction” script on a corpus, what do the C-value, Gfidf, Specificity chi2 values mean?
- After producing a Sankey diagram using the “Network Mapping” script, what do the different colors of the bars mean? Why are some topics the same color?
- After running the “Distant Reading” script on a corpus, what does the “Pertinence” column mean?
- After producing a Bump Graph using the “Epic Epoch” script, what do the different colors mean? Why are some terms the same color?
I was also wondering the best way to cite Cortext and the documentation pages?
Thank you so much!
Dear Kate Li,
- Unithood (CValue), Ranking principal (Gfidf, Specificity) and Chi2 metrics
- In a Sankey diagram, same color refers to the same inter-temporal lexical wave / stream. It also corresponds to the color of the clusters produced for each period (accessible in the pdf and svg files).
- By selecting several windows for the cooccurrences calculation (Size of the window when computing cooccurrences) the pertinence column will be filled by one bump chart and one egonetwork per selected Size of the window (Context window configuration). So, in the bump graphs / egonetworks you will be able to visualize the neighbours that surround the noun phrase (and its composition) written at the beginning of the line;
- In the Bump Chart produced by Epic Epoc, the color does not have a specific meaning except that the color is the same for all the frequencies of a main form over time (or any other categorical variable used with it).
I hope it helps