Update README.md

README.md

@@ -233,45 +233,135 @@ size_categories:
 ---
 
 
-# Dataset Card for CNN Dailymail Dataset
+# Dataset Card for **WikiSQE\_experiment**
 
 ## Dataset Description
 
-- **Repository**: https://github.com/ken-ando/WikiSQE
-- **Paper**: https://arxiv.org/abs/2305.05928
+* **Repository**: [https://github.com/ken-ando/WikiSQE](https://github.com/ken-ando/WikiSQE)
+* **Paper**: [https://arxiv.org/abs/2305.05928](https://arxiv.org/abs/2305.05928) (AAAI 2024)
 
 ### Dataset Summary
 
-[WikiSQE: A Large-Scale Dataset for Sentence Quality Estimation in Wikipedia](https://arxiv.org/abs/2305.05928) by [Kenichiro Ando](https://ken-ando.github.io/kenichiro_ando/index.html), Satoshi Sekine and Mamoru Komachi (AAAI 2024).
+`WikiSQE_experiment` is the **official evaluation split** for **WikiSQE: A Large‑Scale Dataset for Sentence Quality Estimation in Wikipedia**.
 
-The WikiSQE Dataset is an English-language dataset containing over 3.4M sentences in Wikipedia. Dataset's sentences seem to be poor quality in some aspects by Wikipedia editors. The aspects of poor qualities are classified into 153 labels. This repository is a split for experiments used in our paper and contains 5 categories and the top 20 most frequent labels. Each subset contains labeled and unlabeled sentences blended at a 1:1 ratio. Whole dataset is in place https://huggingface.co/datasets/ando55/WikiSQE. 
+While the parent dataset (`ando55/WikiSQE`) contains **every** sentence flagged with a quality problem in the full edit history of English Wikipedia, **this repo provides the exact train/validation/test partitions used in the AAAI 2024 paper**. It offers **≈ 8.3 million sentences** organised as:
 
-A list of category: ['all', 'citation', 'disputed claim', 'information addition', 'other', 'sac', 'syntactic or semantic revision']
+* **27 dataset *groups*** (20 frequent quality labels + 5 Quality type categories + 2 Coarse groups)
+* **3 standard splits per group** (`train`, `val`, `test`) – for example `citation/train`, `citation/val`, …
+
+> **Need the full dump?** Head to [https://huggingface.co/datasets/ando55/WikiSQE](https://huggingface.co/datasets/ando55/WikiSQE).
+
+---
+
+## Dataset Structure
+
+### Groups (27)
+
+| Group                         | List of labels                                                                                                                                                                                                                                                                                                                              |
+| ----------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
+| **Quality type categories** (5)        | ['citation', 'disputed claim', 'information addition', 'other', 'syntactic or semantic revision']                                                                                                                                                                                                                              |
+| **Most‑frequent labels** (20) | ['according to whom', 'attribution needed', 'by whom', 'citation needed', 'clarification needed', 'dead link', 'disambiguation needed', 'dubious', 'needs update', 'neutrality disputed', 'not in citation given', 'original research', 'pronunciation', 'sic', 'unreliable source', 'vague', 'verification needed', 'when', 'which', 'who'] |
+| **Coarse groups** (2)        | ['all', 'sac']                                                                                                                                                                                                                              |
+**Notes**
+
+* **`all`** is a **random subset uniformly sampled from the entire WikiSQE corpus**. Use it when you want a representative slice without downloading the full 3.4 M‑sentence dump.
+* **`sac`** is a **composite set randomly drawn from the three fine‑grained categories `disputed claim`, `information addition`, and `syntactic or semantic revision`**. It was introduced in the paper to study sentence‑level action classification.
+
+### Split sizes
+
+| Split   | Number of sentences |
+| ------- | ------------------------ | 
+| `train` | Depends on labels    |
+| `val`   | 1 k                | 
+| `test`  | 1 k                 |
 
-A list of labels: ['according to whom', 'attribution needed', 'by whom', 'citation needed', 'clarification needed', 'dead link', 'disambiguation needed', 'dubious', 'needs update', 'neutrality disputed', 'not in citation given', 'original research', 'pronunciation', 'sic', 'unreliable source', 'vague', 'verification needed', 'when', 'which', 'who']
 
 ### Data Fields
 
-- `text`: a string feature
-- `label`: a ClassLabel feature (1: labeled sentence, 0: non-labeled sentence)
+| Field   | Type        | Description                                                                                                                      |
+| ------- | ----------- | -------------------------------------------------------------------------------------------------------------------------------- |
+| `text`  | *string*    | Sentence taken from a specific Wikipedia revision                                                                                |
+| `label` | *int* (0/1) | **1** = sentence is tagged with the current config’s quality issue; **0** = sentence from the same revision **without** that tag |
+
+
+---
+
+## Download & Usage
+
+### 1 — Download the Parquet snapshot
+
+```bash
+# Install (if you haven't already)
+pip install --upgrade datasets huggingface_hub
+```
+
+```python
+from huggingface_hub import snapshot_download
+
+repo_dir = snapshot_download(
+    repo_id="ando55/WikiSQE_experiment",  # this repo
+    repo_type="dataset",
+    local_dir="WikiSQE_experiment_parquet",
+    local_dir_use_symlinks=False,
+)
+print("Saved at:", repo_dir)
+```
+
+This grabs **all 27 configs** (each providing `train`, `val`, `test`) in their native **Parquet** format.
+
+### 2 — Load a split on‑the‑fly
+
+Streaming access without a full download:
+
+```python
+from datasets import load_dataset
+
+ds = load_dataset(
+    "ando55/WikiSQE_experiment",
+    name="citation",   # choose any config
+    split="train",
+    streaming=True
+)
+```
+
+### 3 — (Optionally) Convert Parquet → CSV
 
-### Label Details and Statistics
+```python
+import pyarrow.dataset as ds, pyarrow.csv as pv, pyarrow as pa, pathlib
 
-See https://github.com/ken-ando/WikiSQE.
+src = pathlib.Path("WikiSQE_experiment_parquet")
+dst = pathlib.Path("WikiSQE_experiment_csv"); dst.mkdir(exist_ok=True)
 
-### Citation Information
+for pq in src.rglob("*.parquet"):
+    cfg   = pq.parent.name  # config name
+    split = pq.stem         # train/val/test
+    print(cfg, split)
+    out   = dst / f"{cfg}_{split}.csv"
+    first = not out.exists()
+    dset  = ds.dataset(str(pq))
+    with out.open("ab") as f, pv.CSVWriter(
+            f, dset.schema,
+            write_options=pv.WriteOptions(include_header=first)) as w:
+        for batch in dset.to_batches():
+            w.write_table(pa.Table.from_batches([batch]))
 ```
+
+---
+
+## Citation
+
+```bibtex
 @inproceedings{ando-etal-2024-wikisqe,
-    title = "WikiSQE: A Large-Scale Dataset for Sentence Quality Estimation in Wikipedia",
-    author = "Ando, Kenichiro  and
-      Sekine, Satoshi  and
-      Komachi, Mamoru",
-    booktitle = "Proceedings of the AAAI Conference on Artificial Intelligence",
-    volume= "38",
-    number= "16",
-    pages= "17656--17663",
-    year= "2024",
-    address = "Vancouver, Canada",
-    publisher = "Association for the Advancement of Artificial Intelligence",
+  title     = {{WikiSQE}: A Large-Scale Dataset for Sentence Quality Estimation in Wikipedia},
+  author    = {Ando, Kenichiro and Sekine, Satoshi and Komachi, Mamoru},
+  booktitle = {Proceedings of the AAAI Conference on Artificial Intelligence},
+  year      = {2024},
+  volume    = {38},
+  number    = {16},
+  pages     = {17656--17663},
+  address   = {Vancouver, Canada},
+  publisher = {Association for the Advancement of Artificial Intelligence}
 }
-```
+```
+
+*Happy experimenting!* 🚀