Data
This page provides truncated corpora used in class demonstrations and assignments. These are sampled subsets of larger corpora maintained in the NLP Corpora for Korean Studies repository.
Organizing your data folder: We recommend creating subfolders within your /data directory organized by file type or corpus name – for example, /data/president_speeches/, /data/newspapers/, etc. This will keep things tidy as we work with more datasets throughout the semester.
| Week | Corpus | Description | Truncation | Full Corpus | Download |
|---|---|---|---|---|---|
| 2 | Presidential Speeches (Democratic Era) | 749 Korean presidential speeches from the democratic era (6th Republic onward): Roh Tae-woo through Moon Jae-in. Variables include president, title, date, location, speech type, and full text. | Proportionally sampled from 5,840 speeches (random seed 42), with representation across all 7 democratic-era presidents. | Full corpus | CSV (~4.4 MB) |
How to Use These Files
- Download the CSV file from the link above
- Add it to a subfolder in your GitHub repository (e.g.,
/data/president_speeches/) - Commit and push via GitHub Desktop
- Load the file in Orange Data Mining using the Corpus widget
For details on sampling methods and variable descriptions, see the data documentation.