GEMR KG: Ground Truth Questions for AI Generated SPARQL Query Validation

A 31.1 KB answer key for training and testing AI models, published on figshare under a CC-BY-4.0 license. It pairs natural language questions about developing economies with corresponding SPARQL queries for extracting financial risk data from a knowledge graph. The dataset was authored by Adishesh Gonibeed Ravishankar and last updated on 2026-05-30.

Use Cases

• Training AI models for SPARQL query generation based on natural language questions about financial risk.
• Benchmarking the accuracy of AI-generated SPARQL queries against a ground truth.
• Developing question-answering systems for financial knowledge graphs based on the described query-answer pairs.

Strengths

• Provides a specific ground truth for validating AI-generated SPARQL queries.
• Released under a permissive CC-BY-4.0 license for open use.
• Focuses on a niche domain of financial risk data for developing economies.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset is very small at 31.1 KB, indicating a limited scope.

Provenance

Source

figshare

Freshness

Last updated 2026-05-30 22:58:08; freshness should be verified.

Geography

Developing economies