🍽️ Food Recommendation and Personalized Banner Creation System

🌟 Project Overview

This innovative system combines recommendation algorithms, LLM (Large Language Models), and DALL·E 3 to create personalized advertisement banners. The system features:

🔍 Recommendation System: Analyzes user interaction data for personalized product suggestions 🤖 LLM for Text Generation: Creates tailored product descriptions and offers 🎨 DALL·E 3 for Image Creation: Generates visual product representations 🎯 Personalized Banners: Adds eye-catching advertisement banners to images

🛠️ Components Description

1. 📊 Recommendation Algorithm

Two powerful approaches:

• Content-based Filtering: Analyzes user's previous interactions
• Collaborative Filtering: Uses similar users' preferences via Nearest Neighbors models

2. 🧠 LLM (Large Language Model)

Leverages models like AI21 and OpenAI to generate engaging descriptions. Example:

"While you're enjoying tasty apples with other fruits, we recommend complementing it with a delicious juice or light dessert."

3. 🎨 DALL·E 3 Integration

Creates stunning visual representations based on text prompts for marketing purposes.

4. 🎯 Personalized Advertisement Banners

Overlays customized banners with recommendations and promotions on generated images.

⚙️ How It Works

1. 📊 Product Recommendations: Analyzes user history
2. ✍️ Text Generation: Creates personalized marketing messages
3. 🖼️ Image Creation: Generates product visuals with DALL·E 3
4. 🎨 Banner Addition: Overlays personalized offers
5. 💾 Result Saving: Stores final images for campaigns

🚀 Installation

1. Clone the repository:

   git clone https://github.com/MasterPo696/BuyThisToo.git

2. Navigate to the project directory:

   cd BuyThisToo

3. Install the required dependencies:

   pip install -r requirements.txt

4. Set up the necessary paths to data files and API keys as required.

📋 Usage

1. Make sure you have the following data files:

   • user_interactions.csv: Data on user interactions with products.
   • final_standardized_labelled.csv: Data on products with labels.
   • Ensure the files are placed in the correct folders.

2. Run the program:

   python main.py

3. The result will be an image with a banner overlaid, showcasing product recommendations.

📁 Project Structure

BuyThisToo/
│
├── data/
│   ├── user_interactions.csv
│   └── final_standardized_labelled.csv
│
├── app/
│   ├── banner.py
│   ├── image.py
│   ├── processing.py
│   └── __init__.py
│
├── config.py
├── llm/
│   └── ai21.py
├── recomender.py
└── main.py

• data/ — Contains user and product data files.
• app/ — Directory containing the functions for image processing, banner creation, and description cleaning.
• llm/ — Interface for working with AI models like AI21 or OpenAI.
• recomender.py — File that contains the recommendation system logic.
• main.py — The main executable file that runs the entire process.

📝 Notes

• 🔑 Ensure API access for AI21, OpenAI, or other LLMs.
• 📚 Required libraries:
  • PIL for image processing
  • requests for web services
  • scikit-learn for recommendation algorithms

📄 License

This project is licensed under the MIT License.

---

Made with ❤️ for food recommendation and marketing automation