Romify

Discover • Explore • Experience

A modern full-stack accommodation marketplace built with the MERN ecosystem, enabling users to discover unique stays, create listings, share experiences through reviews, and manage properties through a secure and seamless platform.

🚀 Full-Stack Property Rental Platform with AI-Powered Recommendations

About The Project

Romify is a feature-rich accommodation marketplace that allows users to discover unique stays, create and manage property listings, upload images, and share experiences through reviews.

The platform also includes an AI-Powered Smart Stay Recommender that understands natural-language travel requirements and intelligently matches users with the most relevant properties.

Built using modern web technologies and following the MVC architecture, Romify delivers a scalable, secure, and user-friendly accommodation discovery experience.

✨ Key Features

🔐 Authentication & Security

Secure User Registration & Login
Passport.js Authentication
Session Management
Cookie-Based Authentication
Protected Routes & Authorization

🏠 Listing Management

Create New Listings
Edit Existing Listings
Delete Listings
Detailed Listing Pages
Dynamic Image Uploads
Property Capacity & Amenities
Property Ratings

🤖 AI-Powered Smart Stay Recommender

Romify uses AI to understand what users are looking for and recommend properties based on their natural-language travel requirements.

✨ Features

🧠 Natural Language Understanding
- Users can describe their travel plans in plain English.
- Example:
  
  "I'm going to Manali with 3 friends. I need a peaceful place with mountain views, Wi-Fi and parking under ₹5000."
🎯 Intelligent Property Matching
- AI extracts important requirements such as:
  - Location
  - Budget
  - Group size
  - Amenities
  - Property preferences
⭐ Match Scoring
- Every recommendation receives a percentage-based match score.
🏆 Ranked Recommendations
- Properties are automatically sorted from the best match to the least relevant match.
💡 AI Explanations
- Each recommendation includes an explanation describing why the property matches the user's requirements.

📊 Matching Algorithm

The recommendation engine evaluates multiple factors:

Matching Factor	Weight
💰 Budget	25%
📍 Location	25%
👥 Capacity	15%
🛎️ Amenities	20%
✨ Preferences	15%

🧠 AI Technology

Groq API
Mixtral 8x7B
Natural Language Processing
Rule-Based Fallback Parsing
Weighted Recommendation Algorithm

The system also includes fallback parsing using regex patterns, allowing basic recommendation functionality even when the AI API is unavailable.

⭐ Reviews & Feedback

Add Reviews
Delete Reviews
User Interaction System
Property Feedback
Rating-Based Property Evaluation

☁️ Cloud Integration

Cloudinary Image Storage
Optimized Media Delivery
Secure Image Management
Scalable Cloud Infrastructure

🎨 User Experience

Responsive EJS Templates
Flash Messages
Clean UI Design
Mobile-Friendly Layout
Modern AI Recommendation Interface

🛠️ Tech Stack

Frontend	HTML5, CSS3, JavaScript, EJS
Backend	Node.js, Express.js
Database	MongoDB, Mongoose
Authentication	Passport.js, Express Session, Cookies
AI / NLP	Groq API, Mixtral 8x7B, Axios
Cloud Services	Cloudinary
Architecture	MVC Pattern

🏗️ Project Architecture

                         ┌─────────────────────┐
                         │       Client        │
                         │     EJS + JS        │
                         └──────────┬──────────┘
                                    │
                                    ▼
                         ┌─────────────────────┐
                         │       Routes       │
                         └──────────┬──────────┘
                                    │
                    ┌───────────────┴───────────────┐
                    ▼                               ▼
          ┌─────────────────┐             ┌──────────────────┐
          │   Controllers   │             │ Recommendation   │
          │                 │             │    Controller    │
          └────────┬────────┘             └────────┬─────────┘
                   │                               │
                   │                               ▼
                   │                    ┌──────────────────┐
                   │                    │   Groq AI / NLP  │
                   │                    │     Processing   │
                   │                    └────────┬─────────┘
                   │                             │
                   └──────────────┬──────────────┘
                                  ▼
                         ┌─────────────────────┐
                         │   Models / MongoDB  │
                         └─────────────────────┘

🤖 AI Recommendation Flow

User Natural Language Query
            │
            ▼
      Groq AI / NLP
            │
            ▼
   Extract Requirements
            │
     ┌──────┼─────────┐
     ▼      ▼         ▼
 Location Budget   Amenities
     │      │         │
     └──────┼─────────┘
            ▼
     Search Listings
            │
            ▼
    Calculate Match Score
            │
            ▼
      Rank Properties
            │
            ▼
   Top 5 Recommendations
            │
            ▼
 Property + Score + Explanation

🎯 AI Recommendation Example

User Query

I'm going to Manali with 3 friends.
We need a peaceful place with mountain views,
Wi-Fi and parking, under ₹5000 per night.

AI Processing

The system extracts:

Location     → Manali
Budget       → ₹5000/night
Group Size   → 3-4 people
Amenities    → Wi-Fi, Parking
Preferences  → Mountain Views, Peaceful

Example Recommendation

🏆 Mountain View Villa

Match Score: 94%

✓ Within your budget
✓ Suitable for your group
✓ Located in Manali
✓ Wi-Fi available
✓ Parking available
✓ Strong match for mountain-view preference

AI Explanation:
"Excellent match for your budget and group size.
The property provides Wi-Fi and parking and strongly
matches your preference for a peaceful mountain-view stay."

🔗 AI Recommendation API

GET `/recommendations`

Displays the AI recommendation interface.

POST `/recommendations/search`

Processes a natural-language travel query.

Request

{
  "query": "I want a villa in Goa with pool under ₹8000"
}

Response

{
  "success": true,
  "recommendations": [
    {
      "_id": "...",
      "title": "Beachfront Resort in Goa",
      "location": "Goa",
      "price": 3500,
      "capacity": 2,
      "amenities": [
        "Wi-Fi",
        "Beach Access",
        "Pool"
      ],
      "rating": 4.6,
      "matchScore": 92,
      "explanation": "Perfect match for your budget, features the pool you requested, and has excellent beach access in Goa."
    }
  ]
}

🗂️ AI Recommendation Data Model

For the AI recommender to work effectively, listings support additional information such as:

{
  title: "Mountain View Villa",
  description: "Peaceful villa surrounded by mountains",
  price: 4200,
  location: "Manali",
  country: "India",
  capacity: 4,
  amenities: [
    "Wi-Fi",
    "Parking",
    "Kitchen"
  ],
  rating: 4.8
}

These attributes are used by the recommendation engine to calculate property relevance.

🚀 Installation

Clone the Repository

git clone https://github.com/your-username/romify.git

Navigate to the Project Directory

cd romify

Install Dependencies

npm install

This installs all required dependencies including Axios for AI API communication.

🔑 Environment Variables

Create a .env file in the project root:

ATLASDB_URL=your_mongodb_connection_string

CLOUD_NAME=your_cloudinary_name
CLOUD_API_KEY=your_cloudinary_api_key
CLOUD_API_SECRET=your_cloudinary_api_secret

SECRET=your_session_secret

GROQ_API_KEY=your_groq_api_key

NODE_ENV=development

Get a Groq API Key

Create a free Groq API key from:

Groq API Console

Never commit your .env file or API keys to GitHub.

▶️ Run the Application

node app.js

or

npm start

Then open:

http://localhost:8080

For the AI recommendation interface:

http://localhost:8080/recommendations

📂 Project Structure

Romify/
│
├── controllers/
│   └── ...
│
├── controler/
│   └── recommendation.js
│
├── models/
│   └── listing.js
│
├── routes/
│   └── ...
│
├── middleware/
│   └── ...
│
├── utils/
│   └── recommendationAI.js
│
├── public/
│   ├── css/
│   ├── js/
│   └── images/
│
├── views/
│   ├── layouts/
│   ├── listings/
│   ├── recommendations/
│   │   └── index.ejs
│   ├── users/
│   └── includes/
│
├── init/
│   └── data.js
│
├── app.js
├── package.json
├── .env.example
└── README.md

🎯 Key Highlights

✅ Full-Stack MERN Ecosystem

✅ AI-Powered Natural Language Recommendations

✅ Intelligent Property Matching

✅ Match Score & AI Explanations

✅ Groq + Mixtral AI Integration

✅ Secure Authentication System

✅ RESTful API Design

✅ Cloud-Based Image Management

✅ Responsive EJS Interface

✅ Scalable MVC Architecture

✅ MongoDB Database Integration

✅ Flash Messaging System

📸 Screenshots

🏠 Homepage

🏡 Listings

📁 Recommended Screenshot Structure

screenshots/
│
├── homepage.png
├── listings.png
├── property-details.png
├── login.png
├── reviews.png
└── ai-recommendations.png

📚 Learning Outcomes

Through Romify, the project demonstrates practical experience in:

Full-Stack Web Development
RESTful API Development
Natural Language Processing
AI API Integration
Recommendation Systems
Weighted Matching Algorithms
Authentication & Authorization
MongoDB Data Modeling
Cloudinary Integration
Session & Cookie Management
MVC Architecture
Responsive UI Development
Scalable Application Design

🔮 Future Enhancements

🔍 Advanced Search & Filters
🗺️ Interactive Maps
❤️ Wishlist Feature
💳 Payment Integration
📱 Progressive Web App
🔔 Real-Time Notifications
🌐 Multi-Language Support
🧠 Personalized Recommendations using User History
📊 Recommendation Analytics
🏡 Similar Property Recommendations
⭐ Recommendation Learning from User Feedback
🕒 Recommendation History

👨‍💻 Developer

Sumit Kumar

B.Tech Undergraduate | NIT Kurukshetra

Full Stack Developer Open Source Contributor

⭐ If you like this project, consider giving it a star!

Built with ❤️ using Node.js, Express.js, MongoDB, EJS, Passport.js, Cloudinary & Groq AI

Name		Name	Last commit message	Last commit date
controler		controler
init		init
models		models
public		public
utils		utils
views		views
.ENV		.ENV
.gitignore		.gitignore
README.md		README.md
app.js		app.js
package-lock.json		package-lock.json
package.json		package.json

Latest commit

History

Folders and files

Repository files navigation