SubGTR

This is the code for the manuscript "Temporal Knowledge Graph Question Answering via Subgraph Reasoning" (KBS). Paper: https://www.sciencedirect.com/science/article/pii/S0950705122005603

Installation

Clone and create a conda environment

git clone https://github.com/czy1999/SubGTR.git

The implementation is based on TempoQR in TempoQR: Temporal Question Reasoning over Knowledge Graphs and their code from https://github.com/cmavro/TempoQR. You can find more installation details there. We use TComplEx KG Embeddings as implemented in https://github.com/facebookresearch/tkbc.

Dataset and pretrained models download

Complex-CronQuestions can be found in ./ComplexCronQuestions folder.

For CronQueestions: Download and unzip data.zip and models.zip in the root directory.

Drive: https://drive.google.com/drive/folders/1aS2s5sZ0qlDpGZ9rdR7HcHym23N3pUea?usp=sharing.

Running the code

SubGTR on CronQuestions:

python ./train_qa_model.py --model subgtr --subgraph_reasoning --time_sensitivity --aware_module

SubGTR on Complex-CronQuestions (create the wikidata_big_complex folder in advance ):

python ./train_qa_model.py --model subgtr --dataset_name wikidata_big_complex --subgraph_reasoning --time_sensitivity --aware_module

Please explore more argument options in train_qa_model.py.

Note：Score Fusion module will be released soon.

Cite

If you find our method, code, or experimental setups useful, please cite our paper:

@article{DBLP:journals/kbs/ChenZLLK22,
  author    = {Ziyang Chen and
               Xiang Zhao and
               Jinzhi Liao and
               Xinyi Li and
               Evangelos Kanoulas},
  title     = {Temporal knowledge graph question answering via subgraph reasoning},
  journal   = {Knowl. Based Syst.},
  volume    = {251},
  pages     = {109134},
  year      = {2022},
}