Update README.md

README.md

@@ -1,19 +1,32 @@
 ---
 dataset_name: grammar-correction
 description: >
-  A refined subset of the liweili/c4_200m dataset, derived from Google's C4_200M Synthetic Dataset for Grammatical Error Correction. 
-  Contains sentence pairs where the input is ungrammatical and the output is grammatical, suitable for training grammatical error correction (GEC) models.
+  A refined subset of the liweili/c4_200m dataset, derived from Google's C4_200M
+  Synthetic Dataset for Grammatical Error Correction.  Contains sentence pairs
+  where the input is ungrammatical and the output is grammatical, suitable for
+  training grammatical error correction (GEC) models.
 train_size: 100000
 validation_size: 25000
 data_fields:
-  - input: "Ungrammatical sentence (string)"
-  - output: "Corrected, grammatical sentence (string)"
+- input: Ungrammatical sentence (string)
+- output: Corrected, grammatical sentence (string)
 intended_uses: >
-  Designed for training and evaluating GEC models to automatically identify and correct grammatical errors in text.
+  Designed for training and evaluating GEC models to automatically identify and
+  correct grammatical errors in text.
 limitations:
-  - "Synthetic data may not fully capture real-world grammatical errors."
-  - "Filtering may introduce biases based on classifier performance."
-  - "Focused on English grammar, limiting applicability to other languages or dialects."
+- Synthetic data may not fully capture real-world grammatical errors.
+- Filtering may introduce biases based on classifier performance.
+- >-
+  Focused on English grammar, limiting applicability to other languages or
+  dialects.
+task_categories:
+- text2text-generation
+- text-classification
+language:
+- en
+tags:
+- gec
+- grammar
 ---
 # grammar-correction