# xiangwang1223/knowledge_graph_attention_network

KGAT: Knowledge Graph Attention Network for Recommendation, KDD2019

Repository: https://github.com/xiangwang1223/knowledge_graph_attention_network
Canonical: https://ross.abutalabs.com/products/knowledge_graph_attention_network
Language: Python
License: MIT
License Family: permissive
Topics: recommender-system, knowledge-graph, graph-attention-networks, graph-neural-networks, kdd2019, knowledge-based-recommendation, knowledge-aware-recommendation, explainable-recommendation, embedding-propagation, high-order-connectivity, knowledge-graph-dataset, knowledge-graph-for-recommendation
Last push: 2020-08-05T02:30:32+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2683, "days_push": 2220, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1164, forks 333 (observed 2026-08-28T04:03:49.858142+00:00)

## What it is
TensorFlow implementation of KGAT (Knowledge Graph Attention Network), a KDD 2019 paper on knowledge-aware personalized recommendation. It models high-order connectivity in collaborative knowledge graphs using graph attention networks to improve recommendation with item side information.

## Use cases
- reproduce KGAT results from the KDD 2019 paper
- build a knowledge-graph-aware recommender system
- experiment with graph attention networks for recommendation
- train recommenders on Yelp2018, Amazon-book, or Last-fm datasets
- study embedding propagation over collaborative knowledge graphs
- compare knowledge-aware recommendation baselines

## When to choose
- you need a research baseline for knowledge-aware recommendation
- you want to reproduce or extend the KGAT paper
- your dataset includes a knowledge graph with item relations
- you are studying high-order connectivity in recommender systems

## When to avoid
- you need a production recommender system with active maintenance
- you require modern TensorFlow 2.x or PyTorch support
- you want a plug-and-play recommendation library rather than research code
- you have no knowledge graph available for your items

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning
- domain: machine-learning
- platform: python
- tags: recommender-system, knowledge-graph, graph-neural-networks, graph-attention-networks, kdd2019, tensorflow, collaborative-filtering, research-code, recommender-systems, knowledge-graphs

## Member repositories
- xiangwang1223/knowledge_graph_attention_network (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:49.858142+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:31:31.714976+00:00, confidence not recorded.
  - readme: https://github.com/xiangwang1223/knowledge_graph_attention_network (fetched 2026-08-28T04:03:49.858142+00:00, sha 8ee10a5201e1)
- Data as of 2026-08-30T08:39:29.467469+00:00.