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.
Docs2KG: A Human-LLM Collaborative Approach to Unified Knowledge Graph Construction from Heterogeneous Documents
| Date | Stars |
|---|---|
| 2026-07-31 | 371 |
| 2026-08-03 | 371 |
| 2026-08-06 | 371 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
# Docs2KG
**A Human-LLM Collaborative Approach to Unified Knowledge Graph Construction from Heterogeneous Documents**
[](./docs/files/Docs2KG.v2.pdf)
[](https://docs2kg.ai4wa.com/Video/)
[](https://doi.org/10.5281/zenodo.14788362)
## Installation
We have published the package to PyPi: [Docs2KG](https://pypi.org/project/Docs2KG/),
You can install it via:
```bash
pip install Docs2KG
python -m spacy download en_core_web_sm
```
---
---
## Setup and Tutorial
Detailed setup and tutorial can be found in the [documentation](https://docs2kg.ai4wa.com/Tutorial/1.GettingStarted/).
You have two ways to run the package:
- import the package in the code, and hook it with your own code
- run the package in the command line
### Command Line
```bash
# first setup the CONFIG_FILE environment variable to local one
export CONFIG_FILE=config.yml # or any other path for the configuration file
docs2kg # this command will tell you how to use the package
# we currently support the following commands
docs2kg process-document your_input_file --agent-name phi3.5 --agent-type ollama --project-id your_project_id
docs2kg batch-process your_input_dir --agent-name phi3.5 --agent-type ollama --project-id your_project_id
docs2kg list-formats # list all the supported formats
```
```text
Usage: docs2kg [OPTIONS] COMMAND [ARGS]...
Docs2KG - Document to Knowledge Graph conversion tool.
Supports multiple document formats: PDF, DOCX, HTML, and EPUB.
Options:
-c, --config PATH Path to the configuration file (default: ./config.yml)
--help Show this message and exit.
Commands:
batch-process Process all supported documents in a directory.
list-formats List all supported document formats.
neo4j Load data to Neo4j database.
process-document Process a single document file.
```
```text
Usage: docs2kg process-document [OPTIONS] FILE_PATH
Process a single document file.
FILE_PATH: Path to the document file (PDF, DOCX, HTML, or EPUB)
Options:
-p, --project-id TEXT Project ID for the knowledge graph construction
-n, --agent-name TEXT Name of the agent to use for NER extraction
-t, --agent-type TEXT Type of the agent to use for NER extraction
--help Show this message and exit.
```
```text
Usage: docs2kg neo4j [OPTIONS] PROJECT_ID
Load data to Neo4j database.
Options:
-m, --mode [import|export|load|docker_start|docker_stop]
Mode of operation (import or export)
-u, --neo4j-uri TEXT URI for the Neo4j database
-U, --neo4j-user TEXT Username for the Neo4j database
-P, --neo4j-password TEXT Password for the Neo4j database
-r, --reset_db Reset the database before loading data
--help
```
## Motivation
To digest diverse unstructured documents into a unified knowledge graph, there are two main challenges:
- **How to get the documents to be digitized?**
- With the dual-path data processing
- For image based documents, like scanned PDF, images, etc., we can process them through the layout analysis and
OCR, etc. Docling and MinerU are focusing on this part.
- For native digital documents, like ebook, docx, html, etc., we can process them through the programming parser
- It is promising that we will have a robust solution soon.
- **How to construct a high-quality unified knowledge graph with less effort?**
For now, a lot of tools are focusing on the first challenge, however, overlook the second challeExcerpt of 6,241 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:dc4ccd815b323950, desc:knowledge graph