Dabbu19/yago-4.5-v1

---
license: cc-by-sa-3.0
task_categories:
- text-generation
- feature-extraction
language:
- en
tags:
- rdf
- knowledge-graph
- semantic-web
- triples
size_categories:
- 100K<n<1M
---

# YAGO 4.5

## Dataset Description

Knowledge base combining Wikipedia, WordNet, and GeoNames

**Original Source:** https://yago-knowledge.org/data/yago4.5/yago-4.5.0.2-tiny.zip

### Dataset Summary

This dataset contains RDF triples from YAGO 4.5 converted to HuggingFace
dataset format for easy use in machine learning pipelines.

- **Format:** Originally turtle, converted to HuggingFace Dataset
- **Size:** 10.0 GB (extracted)
- **Entities:** ~2.5M
- **Triples:** ~50M
- **Original License:** CC BY-SA 3.0

### Recommended Use

Balanced size/quality, general knowledge, initial training

### Notes

YAGO 4.5 tiny version. Good balance for initial experiments. ZIP distribution.

## RDF Format

This dataset uses a standard lossless format for representing RDF triples. Each triple is a row with 6 fields:
- `subject`: Subject URI or blank node
- `predicate`: Predicate URI
- `object`: Object value (URI, literal, or blank node)
- `object_type`: Type of object (`uri`, `literal`, or `blank_node`)
- `object_datatype`: XSD datatype URI (for typed literals)
- `object_language`: Language tag (for language-tagged literals)

### Loading the Dataset

```python
from datasets import load_dataset

dataset = load_dataset("yago-4.5")
for row in dataset["train"]:
print(f"{row['subject']} {row['predicate']} {row['object']}")
```

## Citation

If you use this dataset, please cite the original source:

**Dataset:** YAGO 4.5
**URL:** https://yago-knowledge.org/data/yago4.5/yago-4.5.0.2-tiny.zip
**License:** CC BY-SA 3.0

## Conversion Details

- **Converted using:** [RDF to HuggingFace Incremental Converter](https://github.com/CleverThis/cleverernie)
- **Conversion date:** 2026-01-13
- **Format version:** 1.0

---

This dataset is part of the CleverThis knowledge graph collection.