Dataset for paper Graph Visualizations in Prompts for Knowledge Graph Question Answering
Authors/Creators
Description
# Supplementary Dataset
This repository contains the supplementary material for the article:
"Graph Visualizations in Prompts for Knowledge Graph Question Answering"
The dataset contains Knowledge Graph Question Answering (KGQA) instances with ground-truth answers, reasoning paths, entity information, and predictions from multiple models under different prompting configurations, including text-only and multimodal settings.
Additionally, the repository includes the generated graph visualizations used in the experiments. These visualizations are provided in the `graph_visualizations.zip` archive. They are separated by folders for each of the visualization modes. Each file is named according to the id of a row from df_kgqa.pkl.
All files have been anonymized to preserve double-blind review requirements.
## File Structure
- df_kgqa.pkl # KGQA dataset with model predictions
- README.md # Dataset description and documentation
- graph_visualizations.zip # Graph visualizations used in the experiments in .png
The dataset is provided in Python pickle format to preserve structured data fields.
part of the data here contained originates from