daakg's Introduction

DAAKG

Deep Active Alignment of Knowledge Graph Entities and Schemata, SIGMOD 2023

Dataset

Link: DWY100K with dangling entities and schema alignment

Dependencies

Python 3.9+
Python libraries
- PyTorch: conda install pytorch pytorch-cuda=11.8 -c pytorch -c nvidia
- PyG: conda install pyg -c pyg
- graph-tool: conda install -c conda-forge graph-tool
- Other libraries: pip install cupy-cuda11x numpy tensorboard igraph pandas

Quick Start

Running script:

python -m daakg.run --log transe \
    --data_dir "data/daakg/D_W_100K_V1" \
    --save "output" \
    --rate 0.3 \
    --epoch 1000 \
    --check 10 \
    --update 10 \
    --train_batch_size 1024 \
    --share \
    --encoder "" \
    --hiddens "100" \
    --decoder "TransE" \
    --sampling "T" \
    --k "5" \
    --margin "1" \
    --alpha "1" \
    --feat_drop 0.0 \
    --lr 0.01 \
    --train_dist "euclidean" \
    --test_dist "euclidean"

Citation

If you find our work useful, please kindly cite it as follows:

@article{DAAKG_SIGMOD2023,
  author    = { Huang, Jiacheng and Sun, Zequn and Chen, Qijin and Xu, Xiaozhou and Ren, Weijun and Hu, Wei },
  title     = { Deep Active Alignment of Knowledge Graph Entities and Schemata },
  journal   = { Proc. ACM Manag. Data },
  year      = 2023,
  pages     = 159,
  volume    = 1,
  number    = 2
}

daakg's People