Update README.md
Browse files
README.md
CHANGED
|
@@ -10,4 +10,45 @@ pinned: false
|
|
| 10 |
short_description: AI that analyzes restaurant reviews, providing insights
|
| 11 |
---
|
| 12 |
|
| 13 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
short_description: AI that analyzes restaurant reviews, providing insights
|
| 11 |
---
|
| 12 |
|
| 13 |
+
# π½οΈ Restaurant Review Analyzer
|
| 14 |
+
|
| 15 |
+
## Overview
|
| 16 |
+
Restaurant Review Analyzer is an intelligent Streamlit application that transforms customer feedback into actionable insights using advanced machine learning and AI technologies.
|
| 17 |
+
|
| 18 |
+
## Features
|
| 19 |
+
- π Zero-shot sentiment classification across multiple restaurant aspects
|
| 20 |
+
- π€ AI-powered insights generation
|
| 21 |
+
- π Detailed review analysis and visualization
|
| 22 |
+
- π Easy-to-use CSV file upload interface
|
| 23 |
+
|
| 24 |
+
## Prerequisites
|
| 25 |
+
- Python 3.8+
|
| 26 |
+
- Groq API Key
|
| 27 |
+
|
| 28 |
+
## Installation
|
| 29 |
+
|
| 30 |
+
1. Clone the repository:
|
| 31 |
+
2. Create a virtual environment:
|
| 32 |
+
3. Install dependencies:
|
| 33 |
+
4. Set up Groq API Key:
|
| 34 |
+
- Replace `"groq_api_key"` in `app.py` with your actual Groq API key
|
| 35 |
+
|
| 36 |
+
## Usage
|
| 37 |
+
Run the Streamlit application:
|
| 38 |
+
```bash
|
| 39 |
+
streamlit run app.py
|
| 40 |
+
```
|
| 41 |
+
|
| 42 |
+
## CSV File Requirements
|
| 43 |
+
- Must contain a 'Review' column
|
| 44 |
+
- Recommended to have 30 or fewer reviews for initial analysis
|
| 45 |
+
|
| 46 |
+
## Technologies Used
|
| 47 |
+
- Streamlit
|
| 48 |
+
- Transformers (Hugging Face)
|
| 49 |
+
- Pandas
|
| 50 |
+
- Groq AI
|
| 51 |
+
- PyTorch
|
| 52 |
+
|
| 53 |
+
## Contributing
|
| 54 |
+
Pull requests are welcome. For major changes, please open an issue first to discuss proposed changes.
|