Upload README.md with huggingface_hub

README.md

@@ -0,0 +1,68 @@
+---
+title: Romansh–German Parallel Dataset (FineWeb-based)
+colorFrom: gray
+colorTo: red
+sdk: static
+---
+
+# Romansh–German Parallel Dataset (FineWeb-Based)
+
+This dataset contains automatically aligned Romansh–German document pairs, extracted from the [Fineweb2](https://huggingface.co/datasets/HuggingFaceFW/fineweb-2) using cosine similarity over OpenAI embeddings. It was created as part of a university programming project focused on document-level parallel data extraction.
+
+## Contents
+
+The dataset consists of several alignment outputs based on different cosine similarity thresholds and sentence-length penalties:
+
+threshold_matches/
+├── matched_{THRESHOL*100}_{ALPHA*100}.jsonl
+
+├── matched_60_5.0.jsonl
+
+...
+
+
+Each file contains one JSON line per aligned document.
+
+## Format
+
+Each `.jsonl` file contains entries with the following fields:
+
+```json
+{
+  "romansh_text": "Romansh document text",
+  "german_text": "Most similar German document text or null",
+  "similarity": 0.6123,              // Score after optional length penalty
+  "original_similarity": 0.6085       // Raw cosine similarity
+}
+```
+
+## Methodology
+
+1. Documents were embedded using [OpenAI](https://platform.openai.com/docs/models/text-embedding-3-small) `text-embedding-3-small`.
+2. Cosine similarity was computed between Romansh and German documents.
+3. A length penalty was applied:  
+   `penalized_score = cosine_similarity - alpha × |len_diff|`
+4. For each Romansh document, the best-scoring German document was selected.
+5. Only matches above a given threshold were retained.
+
+**Thresholds tested:** `0.5 – 0.8`  
+**Final dataset:** `matched_0.600.jsonl` (best match)
+
+
+## Evaluation
+
+A negative gold standard was used to ensure quality: a set of Romansh documents known to have no valid German alignment. These were checked for false positives.
+
+Unit tests confirmed:
+
+No false matches for known non-aligned documents
+
+Low similarity scores for false-alignment candidates
+
+## File Organization
+
+| File                  | Description                        |
+| --------------------- | ---------------------------------- |
+| `parallel_data/`  | All alignment outputs by threshold |
+| `matched_0.600.jsonl` | Main result (highest quality)      |
+| `README.md`           | You are here                       |