Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
KGAT: Knowledge Graph Attention Network for Recommendation, KDD2019
| Date | Stars |
|---|---|
| 2026-07-31 | 1162 |
| 2026-08-06 | 1161 |
Today
-1 stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
# Knowledge Graph Attention Network
This is our Tensorflow implementation for the paper:
>Xiang Wang, Xiangnan He, Yixin Cao, Meng Liu and Tat-Seng Chua (2019). KGAT: Knowledge Graph Attention Network for Recommendation. [Paper in ACM DL](https://dl.acm.org/authorize.cfm?key=N688414) or [Paper in arXiv](https://arxiv.org/abs/1905.07854). In KDD'19, Anchorage, Alaska, USA, August 4-8, 2019.
Author: Dr. Xiang Wang (xiangwang at u.nus.edu)
## Introduction
Knowledge Graph Attention Network (KGAT) is a new recommendation framework tailored to knowledge-aware personalized recommendation. Built upon the graph neural network framework, KGAT explicitly models the high-order relations in collaborative knowledge graph to provide better recommendation with item side information.
## Citation
If you want to use our codes and datasets in your research, please cite:
```
@inproceedings{KGAT19,
author = {Xiang Wang and
Xiangnan He and
Yixin Cao and
Meng Liu and
Tat{-}Seng Chua},
title = {{KGAT:} Knowledge Graph Attention Network for Recommendation},
booktitle = {{KDD}},
pages = {950--958},
year = {2019}
}
```
## Environment Requirement
The code has been tested running under Python 3.6.5. The required packages are as follows:
* tensorflow == 1.12.0
* numpy == 1.15.4
* scipy == 1.1.0
* sklearn == 0.20.0
## Reproducibility & Example to Run the Codes
To demonstrate the reproducibility of the best performance reported in our paper and faciliate researchers to track whether the model status is consistent with ours, we provide the best parameter settings (might be different for the custormized datasets) in the scripts, and provide [the log for our trainings](https://github.com/xiangwang1223/knowledge_graph_attention_network/tree/master/Log).
The instruction of commands has been clearly stated in the codes (see the parser function in Model/utility/parser.py).
* Yelp2018 dataset
```
python Main.py --model_type kgat --alg_type bi --dataset yelp2018 --regs [1e-5,1e-5] --layer_size [64,32,16] --embed_size 64 --lr 0.0001 --epoch 1000 --verbose 50 --save_flag 1 --pretrain -1 --batch_size 1024 --node_dropout [0.1] --mess_dropout [0.1,0.1,0.1] --use_att True --use_kge True
```
* Amazon-book dataset
```
python Main.py --model_type kgat --alg_type bi --dataset amazon-book --regs [1e-5,1e-5] --layer_size [64,32,16] --embed_size 64 --lr 0.0001 --epoch 1000 --verbose 50 --save_flag 1 --pretrain -1 --batch_size 1024 --node_dropout [0.1] --mess_dropout [0.1,0.1,0.1] --use_att True --use_kge True
```
* Last-fm dataset
```
python Main.py --model_type kgat --alg_type bi --dataset last-fm --regs [1e-5,1e-5] --layer_size [64,32,16] --embed_size 64 --lr 0.0001 --epoch 1000 --verbose 50 --save_flag 1 --pretrain -1 --batch_size 1024 --node_dropout [0.1] --mess_dropout [0.1,0.1,0.1] --use_att True --use_kge True
```
Some important arguments:
* `model_type`
* It specifies the type of model.
* Here we provide six options, including KGAT and five baseline models:
* `kgat` (by default), proposed in [KGAT: Knowledge Graph Attention Network for Recommendation](https://arxiv.org/abs/1905.07854), KDD2019. Usage: `--model_type kgat`.
* `bprmf`, proposed in [BPR: Bayesian Personalized Ranking from Implicit Feedback](https://dl.acm.org/citation.cfm?id=1795167), UAI2009. Such model only uses user-item interactions. Usage: `--model_type bprmf`.
* `fm`, proposed in [Fast context-aware recommendations with factorization machines](https://dl.acm.org/citation.cfm?id=2010002), SIGIR2011. Usage: `--model_type fm`.
* `nfm`, proposed in [Neural Factorization Machines for Sparse Predictive Analytics](https://dl.acm.org/citation.cfm?id=3080777), SIGIR2017. Usage: `--model_type nfm`.
* `cke`, proposed in [Collaborative Knowledge Base Embedding for Recommender Systems](https://dl.acm.org/citation.cfm?id=2939673), KDD2016. Usage: `--model_type cke`.
* `cfkg`, proposed in [LearnExcerpt of 9,676 characters
Read on GitHub
60
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:91ed526fbdc0fef3, topic:knowledge-graph, name:knowledge graph, desc:knowledge graph