Update README.md

README.md

@@ -1,5 +1,5 @@
 ---
-title: Conceptnet Db
+title: ConceptNet DB (All Languages)
 emoji: 📚
 colorFrom: indigo
 colorTo: purple
@@ -8,6 +8,70 @@ sdk_version: 5.49.1
 app_file: app.py
 pinned: false
 license: cc-by-sa-4.0
+tags:
+  - conceptnet
+  - knowledge-graph
+  - all-languages
+  - multilingual
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# ConceptNet Database Explorer (All Languages)
+
+This Gradio application provides access to the **complete, unfiltered ConceptNet 5.5 knowledge graph** with all 28.3 million nodes across **all languages**.
+
+## Why Use This App?
+
+**Use this application if you:**
+- Need access to **all languages** in ConceptNet (not just the 11 languages in the normalized version)
+- Want to explore connections across any language pair
+- Need comprehensive, unfiltered results from the complete knowledge graph
+- Are doing multilingual research that requires less common languages
+
+**For faster queries with 11 specific languages** (en, fr, it, de, es, ar, fa, grc, he, la, hbo), consider using the normalized version which has optimized performance.
+
+## Features
+
+- **Semantic Profile**: Finds incoming and outgoing relations for a given node across all languages
+- **Query Builder**: Allows for custom queries (start_node, relation, end_node) with flexible filtering
+- **Raw SQL**: Enables direct SQL queries against the complete database for advanced users
+
+## Performance Note
+
+This application queries the original 23.6 GB un-normalized database (`cstr/conceptnet-de-indexed`). Queries may take 30-60+ seconds due to:
+- Text-based operations on 34 million edges
+- Comprehensive search across all languages
+- Un-normalized schema optimized for completeness rather than speed
+
+The trade-off is complete language coverage versus query speed.
+
+## Data Quality
+
+This database contains the complete, unfiltered ConceptNet data. You may see some metadata artifacts (such as POS tags like 'n' or 'v') alongside semantic relationships. This is expected behavior for the comprehensive dataset.
+
+## Source Database
+
+This application runs on top of the complete, un-normalized database:
+- **Dataset**: [cstr/conceptnet-de-indexed](https://huggingface.co/datasets/cstr/conceptnet-de-indexed)
+- **Nodes**: 28.3 million nodes across all ConceptNet languages
+- **Edges**: 34 million edges (unfiltered)
+
+## Licensing and Attribution
+
+This work includes data from ConceptNet 5, which was compiled by the Commonsense Computing Initiative. ConceptNet 5 is freely available under the Creative Commons Attribution-ShareAlike license (CC BY SA 4.0) from http://conceptnet.io.
+
+For a full list of licenses and attributions for included resources such as WordNet, Open Multilingual WordNet, and Wikimedia projects, please see the original dataset card.
+
+## Citation Information
+
+If you use this data in your work, please cite the original ConceptNet 5.5 paper:
+
+```bibtex
+@inproceedings{speer2017conceptnet,
+    author = {Robyn Speer and Joshua Chin and Catherine Havasi},
+    title = {ConceptNet 5.5: An Open Multilingual Graph of General Knowledge},
+    booktitle = {Proceedings of the AAAI Conference on Artificial Intelligence},
+    year = {2017},
+    pages = {4444--4451},
+    url = {http://aaai.org/ocs/index.php/AAAI/AAAI17/paper/view/14972}
+}
+```