Datasets:
Update README.md
Browse files
README.md
CHANGED
@@ -22,8 +22,8 @@ Next, we filtered out any books not categorized as public domain.
|
|
22 |
We retrieved the main text of each book corresponding to every row in the CSV and incorporated it into the `text` field.
|
23 |
|
24 |
## 2. Deduplication
|
25 |
-
We removed entries where the `図書カードURL` (Library card URL) in this CSV did not
|
26 |
-
In addition,
|
27 |
|
28 |
## 3. Cleaning
|
29 |
The data in the `text` field was then cleaned in the following sequence:
|
@@ -39,7 +39,7 @@ The data in the `text` field was then cleaned in the following sequence:
|
|
39 |
|
40 |
# Tips
|
41 |
|
42 |
-
If you prefer to employ only modern Japanese, you can filter
|
43 |
|
44 |
# License
|
45 |
CC BY 4.0
|
|
|
22 |
We retrieved the main text of each book corresponding to every row in the CSV and incorporated it into the `text` field.
|
23 |
|
24 |
## 2. Deduplication
|
25 |
+
We removed entries where the `図書カードURL` (Library card URL) in this CSV did not match with the `作品ID` (Work ID) and `人物ID` (Person ID).
|
26 |
+
In addition, entries with text identical to previously encountered text were discarded.
|
27 |
|
28 |
## 3. Cleaning
|
29 |
The data in the `text` field was then cleaned in the following sequence:
|
|
|
39 |
|
40 |
# Tips
|
41 |
|
42 |
+
If you prefer to employ only modern Japanese, you can filter entries with: `row["meta"]["文字遣い種別"] == "新字新仮名"`.
|
43 |
|
44 |
# License
|
45 |
CC BY 4.0
|