Knowledge-Graph-Visualization-Demo

Archive

Neo4j powered KG visulization

There are 2D-Search and 3D-Graph-View for knowledge graph visualization. KG data (COVID-19 traces data from tencent) csv files are in folder Import2Neo4j.

---

Folder List

• Import2Neo4j: Import data to Neo4j
• Neo4j-3D: Read data from Neo4j and use 3D visualization
• KG-Search-Flask: Flask platform for search
• KG-Search-Django: Django platform for search

---

Folder: Import2Neo4j

Neo4j Configuration

• Database Name：COVID-19
• Username：neo4j
• Password：123456

Steps

1. Put all CSV files into the Import folder shown below

2. CYPHER4csvImport: Input the eight statements on by one in this file on the Neo4j console

3. After importing, you will see:

• Node Label: (EVENT, LOCATION, PATIENT, TOPIC)
• Relationship Types: (EVENT_LOCATION, EVENT_TOPIC, PATIENT_EVENT, PATIENT_LOCATION)

---

Folder: Neo4j-3D

It is based on js using 3d-force-graph, which can run directly.

Preview

index.html Setting

1. Configure the server address (no port number required), Neo4j username, and password in the first red box shown above.
2. The second box is the CYPHER statement, which limits the return number to 20000. The return value is json, and the search box can be added to HTML later.

Noted problem

issue2

---

Folder: KG-Search-Flask

• Provide KG graph and triple lists

• The front-end code and back-end code have been separated.

Requirements

• flask
• py2neo
• neo4j
• cpca
• jieba

Run

cd KG-Search-Flask
unzip static.zip

python app.py

Address: 127.0.0.1:5000

Search

Support 3 disorder keywords (space seperated) at most

Examples

• 1 keyword
  • 江苏苏州
  • 确诊
  • 王某
• 2 keywords
  • 隔离 苏州
  • 苏州 男性
  • 苏州 2月2日
  • 2月2日 王某
• 3 keywords
  • 2月2日 苏州 确诊

JSON API

GET EXAMPLE

• 127.0.0.1:5000/api/2月2日 苏州 确诊

POST EXAMPLE

• 127.0.0.1:5000/api

body data: {"string": "2月2日 苏州 确诊"}

Preview

---

Folder: KG-Search-Django

• Django Version for KG-Search

• Provide KG graph and triple lists

• The front-end code and back-end code are not separated.

• Flask version is recommended.

Run

cd KG-Search-Django/web
unzip static.zip
cd ..

python manage.py runserver

Address: 127.0.0.1:8000

---