Datasets:
Update README.md
Browse files
README.md
CHANGED
@@ -21,7 +21,7 @@ We retrieved the main text of each book corresponding to every row in the CSV an
|
|
21 |
|
22 |
## 2. Deduplication
|
23 |
We removed entries where the `図書カードURL` (Library card URL) in this CSV did not coincide with the `作品ID` (Work ID) and `人物ID` (Person ID).
|
24 |
-
|
25 |
|
26 |
## 3. Cleaning
|
27 |
The data in the `text` field was then cleaned in the following sequence:
|
|
|
21 |
|
22 |
## 2. Deduplication
|
23 |
We removed entries where the `図書カードURL` (Library card URL) in this CSV did not coincide with the `作品ID` (Work ID) and `人物ID` (Person ID).
|
24 |
+
In addition, rows with text identical to previously encountered text were discarded.
|
25 |
|
26 |
## 3. Cleaning
|
27 |
The data in the `text` field was then cleaned in the following sequence:
|