Add comprehensive dataset card

README.md

@@ -1,49 +1,95 @@
----
-dataset_info:
-  features:
-  - name: course_name
-    dtype: string
-  - name: subject_area
-    dtype: string
-  - name: peims_code
-    dtype: string
-  - name: credit_hours
-    dtype: float64
-  - name: grade_levels
-    dtype: string
-  - name: teks_code
-    dtype: string
-  - name: strand_name
-    dtype: string
-  - name: expectation_text
-    dtype: string
-  - name: bloom_level
-    dtype: string
-  - name: tac_reference
-    dtype: string
-  - name: staar_tested
-    dtype: bool
-  - name: instructional_units
-    dtype: string
-  splits:
-  - name: train
-    num_bytes: 91698
-    num_examples: 299
-  - name: validation
-    num_bytes: 19755
-    num_examples: 64
-  - name: test
-    num_bytes: 20716
-    num_examples: 65
-  download_size: 67949
-  dataset_size: 132169
-configs:
-- config_name: default
-  data_files:
-  - split: train
-    path: data/train-*
-  - split: validation
-    path: data/validation-*
-  - split: test
-    path: data/test-*
----
+# CCISD TEKS Alignment Dataset (3-Way Split)
+
+## Dataset Description
+
+This dataset contains the alignment between Clear Creek ISD (CCISD) curriculum and Texas Essential Knowledge and Skills (TEKS) standards. The dataset is split into training, validation, and test sets for machine learning applications.
+
+### Dataset Summary
+
+- **Total Records:** 428 TEKS standards
+- **Train Split:** 299 records (69.9%)
+- **Validation Split:** 64 records (15.0%)
+- **Test Split:** 65 records (15.2%)
+- **Subject Areas:** Mathematics, English Language Arts, Science, Social Studies, and more
+- **Grade Levels:** K-12
+- **Courses:** 25 different courses
+
+## Dataset Structure
+
+### Data Fields
+
+Each record contains the following fields:
+
+- `teks_code` (string): Unique TEKS standard identifier (e.g., "MATH.A1.3.B")
+- `subject_area` (string): Subject area (e.g., "Mathematics", "English Language Arts")
+- `course_name` (string): Course name (e.g., "Algebra I", "English I")
+- `grade_level` (string): Grade level or course level
+- `strand_name` (string): Content strand within the subject
+- `expectation_text` (string): Full text of the student expectation
+- `bloom_level` (string): Bloom's Taxonomy level (e.g., "Understand", "Apply", "Analyze")
+
+### Data Splits
+
+The dataset is split using stratified random sampling to ensure balanced distribution:
+
+| Split | Records | Percentage |
+|-------|---------|------------|
+| Train | 299 | 69.9% |
+| Validation | 64 | 15.0% |
+| Test | 65 | 15.2% |
+
+## Usage
+
+```python
+from datasets import load_dataset
+
+# Load the entire dataset
+dataset = load_dataset("robworks-software/ccisd-teks-alignment-split")
+
+# Access individual splits
+train_data = dataset['train']
+val_data = dataset['validation']
+test_data = dataset['test']
+
+# Example: Get first training record
+print(train_data[0])
+```
+
+## Data Source
+
+This dataset is derived from:
+- **Texas Education Agency (TEA)** - Official TEKS standards
+- **Clear Creek Independent School District** - Curriculum alignment data
+
+## Licensing
+
+This dataset is released under the **CC0 1.0 Universal (CC0 1.0) Public Domain Dedication**.
+
+### Attributions
+
+- Texas TEKS standards are government works in the public domain
+- Clear Creek ISD curriculum alignment is publicly available data
+
+## Educational Use Disclaimer
+
+This dataset is provided for educational and research purposes. While derived from authoritative sources, users should verify standards alignment with official TEA documentation for instructional planning.
+
+## Citation
+
+```bibtex
+@dataset{ccisd_teks_alignment_split_2025,
+  title={CCISD TEKS Alignment Dataset (3-Way Split)},
+  author={Robworks Software},
+  year={2025},
+  publisher={HuggingFace},
+  url={https://huggingface.co/datasets/robworks-software/ccisd-teks-alignment-split}
+}
+```
+
+## Dataset Curators
+
+Curated by Robworks Software using official Texas Education Agency TEKS standards and Clear Creek ISD curriculum alignment data.
+
+## Contact
+
+For questions or issues, please contact the dataset maintainers through the HuggingFace dataset repository.