Add README.md

README.md

@@ -1,146 +1,108 @@
 ---
 license: cc-by-sa-4.0
 task_categories:
-- token-classification
+  - token-classification
 task_ids:
-- parsing
-- part-of-speech
-- lemmatization
+  - parsing
+  - part-of-speech
+  - lemmatization
 language:
-- vi
+  - vi
 annotations_creators:
-- machine-generated
+  - machine-generated
 language_creators:
-- found
+  - found
 multilinguality:
-- monolingual
+  - monolingual
 source_datasets:
-- undertheseanlp/UTS_VLC
+  - undertheseanlp/UTS_VLC
 size_categories:
-- 10K<n<100K
+  - 10K<n<100K
 tags:
-- universal-dependencies
-- vietnamese
-- nlp
-- dependency-parsing
-- pos-tagging
-configs:
-- config_name: default
-  data_files:
-  - split: train
-    path: data/train-*
-  - split: validation
-    path: data/validation-*
-  - split: test
-    path: data/test-*
-dataset_info:
-  features:
-  - name: sent_id
-    dtype: string
-  - name: text
-    dtype: string
-  - name: tokens
-    list: string
-  - name: lemmas
-    list: string
-  - name: upos
-    list: string
-  - name: xpos
-    list: string
-  - name: feats
-    list: string
-  - name: head
-    list: string
-  - name: deprel
-    list: string
-  - name: deps
-    list: string
-  - name: misc
-    list: string
-  splits:
-  - name: train
-    num_bytes: 15737093
-    num_examples: 9165
-  - name: validation
-    num_bytes: 747301
-    num_examples: 432
-  - name: test
-    num_bytes: 744334
-    num_examples: 403
-  download_size: 2469001
-  dataset_size: 17228728
+  - universal-dependencies
+  - vietnamese
+  - nlp
+  - dependency-parsing
+  - pos-tagging
 ---
 
 # UDD-1: Universal Dependency Dataset for Vietnamese
 
 ## Dataset Description
 
-Vietnamese Universal Dependency dataset created by Underthesea NLP. This dataset follows the [Universal Dependencies](https://universaldependencies.org/) annotation guidelines.
+Vietnamese Universal Dependency dataset created by Underthesea NLP. This dataset follows the [Universal Dependencies](https://universaldependencies.org/) annotation guidelines and has passed all UD validation levels.
 
 ### Dataset Summary
 
 - **Language:** Vietnamese (vi)
 - **Version:** 1.0
 - **Domain:** ⚖️ Legal (Laws)
-- **Sentences:** 10,000
-- **Tokens:** 230,709
+- **Total Sentences:** 10,000
+- **Total Tokens:** 230,709
 - **Source:** Vietnamese Legal Corpus (UTS_VLC)
 - **Annotation:** Machine-generated using Underthesea NLP toolkit
+- **Validation:** ✅ Passed all UD validation levels (1-5)
+
+### Data Splits
+
+Following [UD guidelines](https://universaldependencies.org/contributing/repository_files.html) for treebanks with >110K tokens:
+
+| Split | Sentences | Tokens | Percentage |
+|-------|----------:|-------:|-----------:|
+| Train | 9,165 | 210,668 | 91.3% |
+| Dev | 432 | 10,033 | 4.3% |
+| Test | 403 | 10,008 | 4.3% |
+| **Total** | **10,000** | **230,709** | **100%** |
 
 ### Statistics
 
-| Metric | Value |
-|--------|-------|
-| Sentences | 10,000 |
-| Tokens | 230,709 |
-| Avg sentence length | 23.07 |
-| Min sentence length | 3 |
-| Max sentence length | 84 |
-
-#### UPOS Distribution
-
-| Tag | Count | Percent |
-|-----|------:|--------:|
-| NOUN | 73,953 | 32.05% |
-| VERB | 56,733 | 24.59% |
-| PUNCT | 26,110 | 11.32% |
-| ADP | 22,534 | 9.77% |
-| AUX | 8,444 | 3.66% |
-| ADV | 7,867 | 3.41% |
-| NUM | 7,509 | 3.25% |
-| ADJ | 7,093 | 3.07% |
-| CCONJ | 5,957 | 2.58% |
-| DET | 5,326 | 2.31% |
-| PRON | 3,853 | 1.67% |
-| SCONJ | 3,793 | 1.64% |
-| PROPN | 1,311 | 0.57% |
-| PART | 154 | 0.07% |
-| X | 72 | 0.03% |
-
-#### Top Dependency Relations
-
-| Relation | Count | Percent |
-|----------|------:|--------:|
-| punct | 26,110 | 11.32% |
-| obj | 21,757 | 9.43% |
-| nmod | 20,998 | 9.10% |
-| case | 20,860 | 9.04% |
-| conj | 19,751 | 8.56% |
-| compound | 12,334 | 5.35% |
-| root | 10,000 | 4.33% |
-| nsubj | 9,527 | 4.13% |
-| acl:subj | 9,362 | 4.06% |
-| cc | 5,927 | 2.57% |
+| Metric | Train | Dev | Test |
+|--------|------:|----:|-----:|
+| Sentences | 9,165 | 432 | 403 |
+| Tokens | 210,668 | 10,033 | 10,008 |
+| Avg length | 23.0 | 23.2 | 24.8 |
 
 ## Usage
 
 ```python
 from datasets import load_dataset
 
+# Load all splits
 dataset = load_dataset("undertheseanlp/UDD-1")
-print(dataset["train"][0])
+print(dataset)
+
+# Access specific split
+train_data = dataset["train"]
+dev_data = dataset["validation"]
+test_data = dataset["test"]
+
+# Example sentence
+print(train_data[0])
 ```
 
+### CoNLL-U Format
+
+The dataset is also available in standard CoNLL-U format:
+- `vi_udd-ud-train.conllu`
+- `vi_udd-ud-dev.conllu`
+- `vi_udd-ud-test.conllu`
+
+## Fields
+
+| Field | Description |
+|-------|-------------|
+| `sent_id` | Sentence identifier |
+| `text` | Original sentence text |
+| `tokens` | Word tokens |
+| `lemmas` | Lemmatized forms |
+| `upos` | Universal POS tags |
+| `xpos` | Language-specific POS tags |
+| `feats` | Morphological features |
+| `head` | Head token indices (1-indexed, 0=root) |
+| `deprel` | Dependency relation labels |
+| `deps` | Enhanced dependencies |
+| `misc` | Miscellaneous annotations |
+
 ## Citation
 
 ```bibtex