Update README.md

README.md

@@ -1,51 +1,87 @@
-# 🤖 Context-Aware Thematic Chatbot
+---
+title: Context-Aware Chatbot
+emoji: 🧠
+colorFrom: blue
+colorTo: indigo
+sdk: gradio
+sdk_version: 3.50.2
+app_file: app.py
+pinned: false
+license: apache-2.0
+---
+
+# 🤖 Context-Aware Chatbot using Thematic Roles and Similarity Matching
 
-This is a simple yet intelligent rule-based chatbot that understands user queries by extracting **thematic roles** (like subject, object, and verb) and then finds the most relevant response using **string similarity**. It doesn't just match words — it understands structure.
+This project is a simple yet intelligent chatbot that uses **Natural Language Processing (NLP)** to understand user input and reply contextually. It combines:
+
+- 🧠 **Thematic Role Extraction** using `spaCy` (to identify subjects, objects, and verbs)
+- 🔍 **String Similarity Matching** using `difflib` (to compare user input with a knowledge base)
+- 💬 A custom **conversation dataset** with real-life queries and responses
+- 🌐 A **Gradio interface** deployed on Hugging Face Spaces
 
 ---
 
-## 🧠 How It Works
+## 🚀 Features
+- Understands context using subject, object, and verb extraction
+- Matches queries based on thematic roles rather than plain keywords
+- Provides explanations for how the response was selected
+- Interactive and beginner-friendly web interface
+- Can be updated easily by editing the CSV knowledge base
 
-The chatbot performs the following steps:
+---
 
-1. **Extract Thematic Roles**:
-   - Uses `spaCy` to parse grammar.
-   - Extracts `subject`, `object`, and `verb` from user input.
+## 🧩 How It Works
+1. The user types a sentence (e.g., "Who are you?")
+2. The chatbot extracts thematic roles (subject, object, verb)
+3. It compares these roles with each query in the CSV using string similarity
+4. If the similarity score crosses a threshold, it returns the best matching response
+5. If no match is found, it responds with a fallback message
 
-2. **Compare with Knowledge Base**:
-   - A CSV file (`clean_basic_conversations.csv`) is used as the chatbot's memory.
-   - Each known query is broken into thematic roles.
-   - String similarity is calculated (using `difflib`) between roles.
+---
 
-3. **Return Best Match**:
-   - The response for the closest matched query is returned if its similarity passes a threshold.
-   - An explanation of how the bot arrived at the response is shown on the side.
+## 📂 Files
+- `app.py` – Main application file (Gradio interface + NLP logic)
+- `clean_basic_conversations.csv` – Custom dataset of conversations
+- `requirements.txt` – Python dependencies
+- `README.md` – Project description and Hugging Face config
 
 ---
 
-## 🖥 Interface
+## 📦 Installation (for local testing)
+```bash
+pip install -r requirements.txt
+python app.py
+```
 
-- Built using **Gradio**.
-- Two-column layout:
-  - Left: Chat interface
-  - Right: "How I thought" explanation panel
+---
+
+## 🖼 Example Conversation
+**You:** Who created you?  
+**Bot:** I was created by two AIML students named Rohit and Ankur.
+
+**You:** What do you do?  
+**Bot:** I help answer questions using simple natural language understanding.
 
 ---
 
-## 📁 Files in This Project
+## 🧠 Technologies Used
+- Python 3
+- spaCy (NLP parsing)
+- difflib (string similarity)
+- pandas (data handling)
+- Gradio (interface)
 
-| File                     | Description                                 |
-|--------------------------|---------------------------------------------|
-| `app.py`                 | Gradio interface for the chatbot            |
-| `chatbot_logic.py`       | Contains logic for thematic role extraction and matching |
-| `clean_basic_conversations.csv` | Knowledge base used by the chatbot     |
-| `requirements.txt`       | Required Python libraries                   |
-| `README.md`              | Project documentation                       |
+---
+
+## 📄 License
+This project is released under the **MIT License**. You are free to use, modify, and distribute it for educational or research purposes.
 
 ---
 
-## 🚀 Running Locally
+## 👨🏻‍💻 Created By
+- Rohit Gomes
+- Ankur Halder
 
-```bash
-pip install -r requirements.txt
-python app.py
+---
+
+Enjoy chatting! 🤖