🥗 NutriGuide AI

Smart Diet & Meal Recommendation System

A full-stack AI-powered nutrition platform that provides personalized meal recommendations using machine learning trained on 230K+ real recipes.

Developed by Akanksha — 2nd Year CSE Student

Features • Project Structure • Tech Stack • Installation • API Docs

👩‍💻 About the Developer

Hi! I'm Akanksha, a 2nd year Computer Science Engineering student. I built NutriGuide AI as a personal project to explore full-stack web development combined with machine learning.

This project helped me learn and apply:

Building REST APIs with Node.js & Express
Designing frontend UIs using React and Tailwind CSS
Training and deploying ML models with Python & Scikit-learn
Connecting a NoSQL database (MongoDB) to a real-world application
Integrating a separate ML microservice with a main backend

.

📖 About the Project

NutriGuide AI is an AI-powered nutrition and meal planning web application. It takes your personal health data (age, weight, height, activity level, fitness goal) and uses a machine learning model to give you a personalized daily calorie target and meal recommendations.

The project is divided into 3 services that work together — a React frontend, a Node.js backend, and a Python ML service.

✨ Features

🤖 ML-Powered Calorie Predictions — Gradient Boosting model with 99.87% accuracy, trained on 230K+ real recipes from Kaggle
🔐 Secure Authentication — JWT-based login & registration with role-based access (Admin / User)
👤 Health Profiles — Set your age, weight, height, activity level, and fitness goal
🍽️ 500+ Meal Database — Search and filter meals by calories, protein, dietary preferences (vegan, keto, gluten-free)
📊 Progress Tracking — Log meals daily and visualize your nutrition trends with charts
🛡️ Admin Dashboard — Manage users and the meal database

📁 Project Structure

NutriGuide/
├── backend/                    # Node.js + Express.js Backend API
│   ├── controllers/            # Business logic for each route
│   │   ├── auth.controller.js
│   │   ├── user.controller.js
│   │   ├── meal.controller.js
│   │   ├── nutrition.controller.js
│   │   ├── progress.controller.js
│   │   └── admin.controller.js
│   ├── models/                 # MongoDB Schemas
│   │   ├── User.model.js
│   │   ├── Meal.model.js
│   │   └── Progress.model.js
│   ├── routes/                 # API route definitions
│   │   ├── auth.routes.js
│   │   ├── user.routes.js
│   │   ├── meal.routes.js
│   │   ├── nutrition.routes.js
│   │   ├── progress.routes.js
│   │   └── admin.routes.js
│   ├── middleware/             # Auth, validation, error handling
│   │   ├── auth.middleware.js
│   │   ├── validation.middleware.js
│   │   └── error.middleware.js
│   ├── seeds/                  # Database seed data
│   │   ├── seed.js
│   │   └── meals_seed.json
│   ├── server.js               # Main Express server entry point
│   ├── .env.example
│   └── package.json
│
├── frontend/                   # React + Vite Frontend
│   ├── src/
│   │   ├── components/         # Reusable UI components
│   │   │   ├── Layout.jsx
│   │   │   ├── PrivateRoute.jsx
│   │   │   └── AdminRoute.jsx
│   │   ├── pages/              # All app pages
│   │   │   ├── LandingPage.jsx
│   │   │   ├── LoginPage.jsx
│   │   │   ├── RegisterPage.jsx
│   │   │   ├── DashboardPage.jsx
│   │   │   ├── MealsPage.jsx
│   │   │   ├── ProgressPage.jsx
│   │   │   ├── AnalyticsPage.jsx
│   │   │   ├── ProfilePage.jsx
│   │   │   └── AdminDashboard.jsx
│   │   ├── services/           # API call functions
│   │   ├── store/              # Zustand global state
│   │   ├── utils/              # Axios config & helpers
│   │   ├── App.jsx             # Routes & app setup
│   │   └── main.jsx            # Entry point
│   ├── index.html
│   ├── tailwind.config.js
│   ├── vite.config.js
│   └── package.json
│
├── ml-service/                 # Python Flask ML Service
│   ├── app.py                  # Flask API for predictions
│   ├── train_model.py          # Train the ML model
│   ├── train_model_with_real_data.py
│   ├── train_meal_recommendation.py
│   ├── preprocess_data.py      # Data preprocessing pipeline
│   ├── create_meal_database.py
│   ├── models/                 # Saved trained model files
│   │   ├── meal_index.json
│   │   ├── model_stats.json
│   │   └── recommendation_stats.json
│   ├── requirements.txt
│   └── README.md
│
├── datasets/                   # Raw nutrition datasets (CSV)
│   ├── nutrition.csv
│   └── processed_meals.csv
│
└── README.md

🚀 Technology Stack

Frontend

Library	Purpose
React 18	UI framework
Vite	Build tool & dev server
Tailwind CSS	Styling
React Router DOM	Page routing
Zustand	Global state management
Axios	HTTP requests
Recharts	Data visualization charts
Framer Motion	Animations
React Hook Form	Form handling
React Hot Toast	Notifications

Backend

Library	Purpose
Node.js + Express.js	Server & API
MongoDB + Mongoose	Database & ODM
JWT	Authentication
bcryptjs	Password hashing
Helmet	Security headers
express-rate-limit	Rate limiting
Morgan	HTTP logging
CORS	Cross-origin support

ML Service

Library	Purpose
Python + Flask	ML API server
Scikit-learn	ML models (Gradient Boosting, Random Forest)
Pandas + NumPy	Data processing
Joblib	Model serialization

🤖 Machine Learning Model

The ML model predicts your daily calorie needs based on your personal health data.

Model: Gradient Boosting Regressor
Dataset: Food.com Recipes (Kaggle) — 230,286 recipes
Accuracy: 99.87% (Test R² = 0.9987, MAE = 5.62 calories)

Input Features (14 total)

Feature	Description
Age	User's age in years
Gender	Male / Female
Height	In centimeters
Weight	In kilograms
BMI	Calculated from height & weight
BMR	Basal Metabolic Rate (Mifflin-St Jeor)
Activity Level	Sedentary → Very Active (0–4)
Fitness Goal	Lose / Maintain / Gain (0–2)
+ 6 engineered features	BMI×Age, Weight/Height, BMR×Activity, etc.

Model Comparison

Model	Test MAE	Test R²
Gradient Boosting ✅	5.62 cal	0.9987
Random Forest	6.75 cal	0.9976
Ridge Regression	47.40 cal	0.9897
Linear Regression	47.43 cal	0.9896

🛠️ Installation & Setup

Prerequisites

Node.js v18+
Python 3.8+
MongoDB (local or Atlas)
npm

Step 1 — Backend Setup

cd backend

# Install dependencies
npm install

# Create environment file
cp .env.example .env
# Edit .env: set MONGODB_URI, JWT_SECRET, ML_SERVICE_URL, FRONTEND_URL

# Start backend (runs on port 5000)
npm run dev

Step 2 — ML Service Setup

cd ml-service

# Create virtual environment
python3 -m venv venv
source venv/bin/activate        # Linux/macOS
# OR: venv\Scripts\activate     # Windows

# Install dependencies
pip install -r requirements.txt

# Train the model (first time only)
python train_model.py

# Start ML service (runs on port 5001)
python app.py

Step 3 — Frontend Setup

cd frontend

# Install dependencies
npm install

# Create environment file
cp .env.example .env
# Set: VITE_API_URL=http://localhost:5000/api

# Start frontend (runs on port 5173)
npm run dev

Step 4 — Seed the Database (Optional)

cd backend
npm run seed

This loads 500+ real meals into MongoDB.

All 3 Services Running

Service	URL
Frontend	http://localhost:5173
Backend API	http://localhost:5000
ML Service	http://localhost:5001

🔐 API Documentation

Base URL: http://localhost:5000/api

Auth Routes

Method	Endpoint	Description
POST	`/auth/register`	Create new account
POST	`/auth/login`	Login & get JWT token

User Routes (requires token)

Method	Endpoint	Description
GET	`/users/dashboard`	Get user dashboard data
PUT	`/users/profile`	Update profile info

Nutrition Routes (requires token)

Method	Endpoint	Description
GET	`/nutrition/recommendations`	Get AI calorie & macro targets
POST	`/nutrition/meal-plan`	Generate meal plan for a date

Meal Routes

Method	Endpoint	Description
GET	`/meals`	Search meals (filter by category, calories)
GET	`/meals/search/recommendations`	Get meals matching calorie target

Progress Routes (requires token)

Method	Endpoint	Description
GET	`/progress/today`	Get today's nutrition log
POST	`/progress/log-meal`	Log a meal
PUT	`/progress/update-weight`	Update weight

🗃️ Database Schema

User

{
  name, email, password (hashed),
  age, gender, height, weight, bmi,
  activityLevel, fitnessGoal,
  dailyCalorieTarget,
  macronutrients: { protein, carbs, fats },
  dietaryPreferences: [],
  allergies: [],
  role: "user" | "admin"
}

Meal

{
  name, description, category,
  nutrition: { calories, protein, carbohydrates, fats, fiber, sugar },
  dietaryTags: [],
  allergens: [],
  ingredients: [],
  prepTime, cookTime
}

Progress

{
  user: ObjectId,
  date,
  weight,
  nutrition: { calories, protein, carbohydrates, fats },
  meals: [{ mealId, mealName, servings, calories, consumedAt }],
  compliance: { calorieCompliance, macroCompliance }
}

📄 License

This project is licensed under the MIT License.

Built with ❤️ by Akanksha — 2nd Year CSE

NutriGuide AI — Empowering healthier lives through AI and technology

Name		Name	Last commit message	Last commit date
Latest commit History 6 Commits
.github		.github
backend		backend
datasets		datasets
frontend		frontend
ml-service		ml-service
.gitignore		.gitignore
README.md		README.md

Folders and files

Latest commit

History

Repository files navigation

🥗 NutriGuide AI

👩‍💻 About the Developer

📖 About the Project

✨ Features

📁 Project Structure

🚀 Technology Stack

Frontend

Backend

ML Service

🤖 Machine Learning Model

Input Features (14 total)

Model Comparison

🛠️ Installation & Setup

Prerequisites

Step 1 — Backend Setup

Step 2 — ML Service Setup

Step 3 — Frontend Setup

Step 4 — Seed the Database (Optional)

All 3 Services Running

🔐 API Documentation

Auth Routes

User Routes (requires token)

Nutrition Routes (requires token)

Meal Routes

Progress Routes (requires token)

🗃️ Database Schema

User

Meal

Progress

📄 License

About

Topics

Resources

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Contributors

Uh oh!

Languages

Packages