Update README.md
Browse files
README.md
CHANGED
|
@@ -14,9 +14,11 @@ This dataset is still a work in progress.
|
|
| 14 |
|
| 15 |
## Data Curation
|
| 16 |
|
| 17 |
-
We download the 2004 EOT WARC files and parse the HTML using Trafilatura. We then filter the extracted text by length (minimum of 550 characters) and language
|
| 18 |
|
| 19 |
-
Tables are preserved, but HTML comments are removed. Additional filtering may be applied
|
|
|
|
|
|
|
| 20 |
|
| 21 |
## Stats
|
| 22 |
|
|
|
|
| 14 |
|
| 15 |
## Data Curation
|
| 16 |
|
| 17 |
+
We download the 2004 EOT WARC files and parse the HTML using Trafilatura. We then filter the extracted text by length (minimum of 550 characters) and language while normalizing blank lines.
|
| 18 |
|
| 19 |
+
Tables are preserved, but HTML comments are removed. Additional filtering may be applied after the parsing phase is complete and all data has been uploaded. That dataset will be known as **EOT-2004-Cleaned**.
|
| 20 |
+
|
| 21 |
+
In the future, we also plan to create an education-focused version of this dataset using Hugging Face's Edu Classifier.
|
| 22 |
|
| 23 |
## Stats
|
| 24 |
|