---
title: AI Memory Graph
emoji: 🧠
colorFrom: red
colorTo: red
sdk: streamlit
app_file: main.py
pinned: false
short_description: NLP & Graph-based Multi-User Chat Memory System
---

# 🧠 AI Memory Graph

Welcome to the live demo of the **AI Memory Graph**! 

This application is an AI-powered memory system that extracts structured knowledge from multi-user group chats. It goes beyond simple chatbots by supporting group dialogues, extracting triplets (subject-predicate-object) using Transformer-based NLP models, and visualizing relationships dynamically.

### 🚀 How to Use It?
1. Enter a **Sender** name and a **Message** in the left sidebar.
2. Click **➕ Add Message** to simulate a group chat environment.
3. Click **🔍 Extract Triplets** to run the SpaCy Transformer model and extract structural information.
4. Finally, click **🌐 Show Knowledge Graph** to render the interactive NetworkX memory graph!

### 🛠️ Tech Stack
* **Backend:** Python, FastAPI logic (adapted for Streamlit)
* **NLP Model:** SpaCy (`en_core_web_trf`)
* **Graph Visualization:** NetworkX, PyVis
* **Frontend:** Streamlit

*Developed by Erdem Yavuz Hacisoftaoglu*