zjunlp/MKG_Analogy

[ICLR 2023] Multimodal Analogical Reasoning over Knowledge Graphs

39
/ 100
Emerging

This project helps researchers and data scientists working with complex datasets to discover deeper connections between entities. It takes a knowledge graph containing facts and related images, then identifies analogies—like spotting patterns such as 'country is to capital as company is to headquarters.' The output helps users understand how different entities relate to each other, even across text and visual information.

132 stars. No commits in the last 6 months.

Use this if you need to find implicit or explicit analogical relationships within a multimodal knowledge graph that combines text-based facts and images.

Not ideal if your data lacks both structured knowledge graph information and associated images, or if you only need simple fact retrieval without analogical reasoning.

knowledge-discovery semantic-analysis pattern-recognition multimodal-data data-relationships
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 13 / 25

How are scores calculated?

Stars

132

Forks

12

Language

Python

License

MIT

Last pushed

Jul 28, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/nlp/zjunlp/MKG_Analogy"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.