Skip to content

CoreSheep/emotion-net

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

77 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

🌟 Emotion Net - Multimodal Emotion Knowledge Graph

An interactive web-based visualization platform for exploring multimodal emotion knowledge graphs, connecting sentiment words, facial expressions, social media emojis, and animal expressions.

📺 Demo Video

🌐 Live Demo

Visit the Live Website →

🎨 Interface

Dark Mode Interface Interactive knowledge graph visualization with dark mode

Light Mode Interface Interactive knowledge graph visualization with light mode

✨ Features

  • 🖱️ Drag & Drop: Click and drag any node to reposition it - watch the graph dynamically adjust its layout
  • 🔍 Zoom Control: Scroll to zoom in and out of the graph for detailed exploration
  • 👆 Hover to Explore: Move your mouse over nodes to see which emotion category they belong to
  • 🔗 Click to Connect: Click on any node to highlight its direct connections and relationships
  • 🎭 Expression Images: View emotion representations across three modalities:
    • 😊 Human Face Expressions - Real facial emotion images
    • 😃 Social Media Emojis - Digital emotion representations
    • 🐱 Animal Expressions - Emotion expressions from cats, dogs, and hamsters
  • 🌓 Dark/Light Mode: Toggle between dark and light themes for comfortable viewing

📊 Dataset

The knowledge graph integrates multimodal emotional data representing 7 core emotions:

Emotion English Description
Joy Happiness, pleasure, delight
Good Positive feelings, satisfaction
Anger Frustration, rage, irritation
Sadness Sorrow, grief, melancholy
Fear Anxiety, worry, terror
Disgust Aversion, revulsion, dislike
Surprise Astonishment, amazement, shock

Dataset Sources:

  • Text Emotion Words: 27,466 Chinese emotion words from Dalian University of Technology
  • Facial Expressions: JAFFE dataset with 216 images across 7 emotions
  • Animal Expressions: 84 images (cats, dogs, hamsters) collected and curated
  • Social Media Emojis: 77 emoji images from emojipedia.org

🏆 Academic Recognition

Bachelor's Thesis

  • Topic: "Construction of a Multimodal Knowledge Graph for Fine-Grained Sentiment Analysis" (view the project website)
  • Institution: Northeastern University (NEU), China

China Invention Patent (Granted)

  • Title: Method for Constructing and Presenting a Multimodal Sentiment Knowledge Graph
  • Patent Number: CN ZL202011319237.8
  • Grant Date: September 7, 2021
  • Status: Co-inventor

🛠️ Technology Stack

  • Frontend: HTML5, CSS3, JavaScript
  • Visualization: D3.js force-directed graph
  • UI Framework: Bootstrap
  • Data Storage: Neo4j graph database
  • Ontology Editor: Protégé

📖 Citation

If you use this work in your research, please cite:

@mastersthesis{li2021emotionnet,
  title={Construction of a Multimodal Knowledge Graph for Fine-Grained Sentiment Analysis},
  author={Li, Jiufeng},
  year={2020},
  school={Northeastern University},
  address={Shenyang, China},
  type={Bachelor's Thesis}
}

Patent Citation:

Li, J. et al. (2021). Method for Constructing and Presenting a Multimodal Sentiment Knowledge Graph. 
China Patent No. CN ZL202011319237.8.

This project represents academic research on multimodal sentiment analysis and knowledge graph construction.

About

A simple but cool visualization project of multimodal Emotions-KG based on d3

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors