Datasets:

aksw
/

Text2SPARQL-Raw

Text2Sparql-Raw is a multilingual dataset designed for the task of translating natural language questions into SPARQL queries over the DBpedia knowledge graph. This dataset aggregates and harmonizes four widely used benchmarks in the text-to-SPARQL domain:

QALD (versions 1–9)
LC-QuAD 1.0
Orange/paraqa-sparqltotext
julioc-p/Question-Sparql

It contains questions in both English and Spanish, making it suitable for multilingual and cross-lingual question answering tasks over knowledge graphs.

This is the first version of the dataset.

✅ Supported Tasks and Leaderboards

Text-to-SPARQL generation: Given a natural language question, the task is to generate a syntactically correct and semantically accurate SPARQL query that retrieves the answer from the DBpedia knowledge graph.

🌍 Languages

English (en)
Spanish (es)

📁 Dataset Structure

Each instance in the dataset includes:

question: The question in natural language (English or Spanish)
query: The corresponding SPARQL query
dataset-id: Original dataset source (e.g., QALD-X, LCQUAD, Orange, JULIOC)

🔧 Dataset Creation

The dataset is a curated compilation of four public datasets:

QALD 1-9: https://github.com/ag-sc/QALD
LC-QuAD 1.0: https://github.com/AskNowQA/LC-QuAD
Paraqa-SparqlToText: https://huggingface.co/datasets/Orange/paraqa-sparqltotext
Question-Sparql: https://huggingface.co/datasets/julioc-p/Question-Sparql

The merged dataset aims to provide a unified resource for training multilingual text-to-SPARQL models.

💡 Use Cases

This dataset is intended for training and evaluating models that:

Generate SPARQL from natural language
Support multilingual question answering over knowledge graphs
Learn cross-lingual representations for semantic parsing

📚 Citation

If you use this dataset in your research, please cite:

@misc{text2sparql-raw,
  author = {Marcos Gôlo, Paulo do Carmo, Edgard Marx, Ricardo Marcacini},
  title = {text2sparql-raw: Natural Language Text to SPARQL for DBpedia Dataset},
  year = {2025},
  howpublished = {\url{https://huggingface.co/datasets/aksw/Text2Sparql-Raw}},
}

🪪 Licensing

license: apache-2.0

Check each dataset’s original repository for full licensing terms.

Dataset Card for Text2sparql-Raw

🧾 Dataset Summary