🧱 Bricked Up: Predictability of a LEGO Set Deal

🚀 Project Overview

Bricked Up is a full-stack web app that aggregates and analyzes LEGO deals 🧩 scraped from public sources. Users can browse deals, filter and sort listings, save favorites, and gain insights with interactive price indicators.

🌐 Live Demo: Visit Bricked Up

---

💡 Why This Project?

Bricked Up solves the challenge of finding the best LEGO deals in a user-friendly, responsive, and automated way. By leveraging scraping, APIs, and automation, it ensures LEGO enthusiasts never miss out on a great deal. 🧱✨

---

✨ Features

• 🛒 View Deals: Browse through aggregated LEGO offers.
• 📊 Relevance Score:
  • Each deal is scored based on its popularity, discount, freshness, and resalability metrics.
  • Relevance helps users prioritize the best deals.
• 🔍 Interactive Filters:
  • 🏆 Best Discount
  • 🔥 Hot Deals
  • 📈 Popular Deals
  • Relevance-based sorting
• 📊 Deal Insights:
  • Average and percentile price indicators.
  • Expiration countdown for time-sensitive offers.
• ❤️ Save Favorites: Mark and revisit your favorite deals.
• 🌗 Dark Mode: Toggle between light and dark themes.
• 🔄 Automated Refresh: Deals update daily at 5 AM and 6 PM UTC+2.
• 📱 Responsive Design: Works seamlessly on all devices, with optimized modals and layouts.
• 🛠️ How It Works Accordion: Guides users on searching, sorting, and understanding the scores.

---

🛠️ Technologies Used

• Frontend: HTML, CSS (Bootstrap 5) 🎨, JavaScript ⚡
• Backend: Node.js with Express.js 🚀
• Database: MongoDB Atlas 🗄️
• Web Scraping: Puppeteer 🕷️, Cheerio 🌿
• Deployment: Vercel 🛠️
• Automation: GitHub Actions 🕒

---

📸 Screenshots

A clean, interactive homepage for LEGO enthusiasts.

Seamless switch to dark mode.

Key price insights with visual indicators.

---

📖 Understanding the Relevance Score

The Relevance Score is a calculated metric that helps users identify the best deals. It evaluates:

• Discount: The percentage off the original price.
• Popularity: Based on the number of comments and likes.
• Freshness: How recently the deal was published.
• Resalability: Resale potential based on average resale prices and listing activity.
• Temperature: A deal’s popularity among community users.
• Expiry: Whether the deal is expiring soon. The score ranges from 0% (low relevance) to 100% (high relevance).

📊 Relevance Score Explained

The Relevance Score is a metric (ranging from 0 to 1) used to rank LEGO deals based on their value and appeal. It evaluates multiple factors with assigned weights to provide a comprehensive score.

Relevance Score Formula

$$ \text{Relevance Score} = W_d \cdot S_d + W_p \cdot S_p + W_f \cdot S_f + W_e \cdot S_e + W_h \cdot S_h + W_r \cdot S_r $$

Where:

• $W$: Weight assigned to each factor
• $S$: Scaled score of each factor
• Subscripts:
  • $d$: Discount
  • $p$: Popularity
  • $f$: Freshness
  • $e$: Expiry
  • $h$: Heat
  • $r$: Resalability

Factors Breakdown

• Discount Score ($S_d$): Percentage discount ( $S_d = \min(\frac{\text{Discount}}{100}, 1)$ ).
• Popularity Score ($S_p$): Community engagement ( $S_p = \min(\frac{\text{Comments}}{\text{MAX COMMENTS}}, 1)$ ).
• Freshness Score ($S_f$): Time since publication ( $S_f = \max(1 - \frac{\text{Days}}{\text{MAX AGE DAYS}}, 0)$ ).
• Expiry Score ($S_e$): Penalizes deals expiring soon ( $S_e = 0.5$ if expiring soon, $S_e = 1$ otherwise).
• Heat Score ($S_h$): Based on temperature ( $S_h = \min(\frac{\text{Temperature}}{\text{MAX TEMPERATURE}}, 1)$ ).
• Resalability Score ($S_r$): Combines:
  • Profitability: ( $\max(\frac{\text{Resale Price} - \text{Price}}{\text{Price}}, 0)$ ),
  • Demand: ( $\min(\frac{\text{Resale Listings}}{\text{MAX LISTINGS}}, 1)$ ),
  • Velocity: ( $\min(\frac{\text{Weekly Resales}}{\text{MAX WEEKLY SALES}}, 1)$ ).

Weight Distribution

• Discount: 20%
• Popularity: 20%
• Freshness: 15%
• Expiry: 5%
• Heat: 10%
• Resalability: 30%
  • Profitability: 50%
  • Demand: 30%
  • Velocity: 20%

The Relevance Score provides a quick, data-driven insight into the best LEGO deals available.

---

⚙️ How It Works

1. Data Collection: 🕷️ Deals are scraped from public sources like Dealabs and Vinted.
2. Backend API: 📡 Data is stored in MongoDB Atlas and served through an Express.js API.
3. Scheduled Updates: ⏰ GitHub Actions refresh the data automatically twice a day.
4. Client Rendering: 🌟 The deals are displayed interactively with filtering, sorting, and responsive design.

---

🌐 Live Updates: Automation

The data refreshes automatically:
⏰ Daily at 5 AM and 6 PM UTC+2
Using GitHub Actions to ensure users always get the latest deals.

---

🗂️ Project Structure

bricked-up/
├── client/
│   └── v2/
│       ├── index.html       # Main client HTML file
│       ├── styles.css       # Custom CSS styles
│       ├── portfolio.js     # Client-side logic
│       ├── assets/          # Images and other assets
│       └── utils.js         # Utility functions
│
├── server/
│   ├── api.js               # Main server file (Express routes)
│   ├── refresh_database.js  # Script to refresh MongoDB
│   ├── dealabs.js           # Scraping script for Dealabs
│   ├── vinted.js            # Scraping script for Vinted
│   └── node_modules/        # Installed dependencies
│
├── .github/
│   └── workflows/
│       └── database-refresh.yml  # GitHub Actions for scheduled scraping
│
├── vercel.json              # Vercel deployment configuration
├── package.json             # Dependencies for server and client
└── README.md                # Project documentation

---

👑 Acknowledgments

• Public data sources: Dealabs and Vinted
• Frameworks & Tools: Bootstrap, Puppeteer, Node.js, MongoDB
• Icons: Flaticon

---

📬 Contact

Developed by: Joyce Lapilus
Project Repository: GitHub

For inquiries, feel free to contact via joyce.lapilus@gmail.com.

---

⚠️ Disclaimer

This website aggregates publicly available data for educational and informational purposes only.
🔒 No malicious intent is associated with data scraping. For any concerns, feel free to contact me.

---

🎉 Thank you for visiting Bricked Up! 🧱✨