Scribd
Upload
3 views

Mini Project

The Diet and Food Recommendation System is a personalized application that utilizes Machine Learning to provide tailored dietary suggestions based on users' health metrics like BMI. It offers a user-friendly interface for inputting health data, visualizes nutritional information, and generates customized meal plans while being scalable through Docker deployment. The system aims to empower users to make informed food choices by adapting recommendations to individual dietary preferences and health goals.

Uploaded by

221501071
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
3 views

Mini Project

The Diet and Food Recommendation System is a personalized application that utilizes Machine Learning to provide tailored dietary suggestions based on users' health metrics like BMI. It offers a user-friendly interface for inputting health data, visualizes nutritional information, and generates customized meal plans while being scalable through Docker deployment. The system aims to empower users to make informed food choices by adapting recommendations to individual dietary preferences and health goals.

Uploaded by

221501071
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 12

RAJALAKSHMI ENGINEERING COLLEGE

Department of Artificial Intelligence and Machine


Learning

AI19611 MINI PROJECT

DIET AND FOOD RECOMMENDATION SYSTEM

Project by,
Madhumitha K(221501071)
ABSTRACT

In today’s health-conscious world, personalized dietary recommendations are essential for maintaining
well-being and managing weight. This project presents a smart food recommendation system powered by
Machine Learning that assesses users’ health metrics such as BMI (Body Mass Index) and provides
dietary suggestions tailored to their body type—underweight, normal, or overweight. The system also
visualizes nutritional data using pie charts and suggests healthy recipes and weight management plans.
Unlike static diet plans, this system adapts to each user's unique health profile, making the
recommendations more practical, interactive, and user-focused..
OBJECTIVE

The primary objective of the Diet Recommendation System is to provide personalized, data-driven diet
recommendations to users based on their unique health parameters such as age, weight, and height. The
system aims to leverage a content-based recommendation approach to suggest healthy food options,
promoting a balanced diet tailored to individual needs. The project seeks to empower users to make informed
food choices by offering customized meal plans, and it also allows for custom food recommendations based
on users' nutritional preferences. Additionally, the system is designed to be scalable and efficient, deploying
the application using Docker and Docker-Compose to ensure smooth and consistent performance across
various environments.
PROBLEM SOLUTION

The Diet Recommendation System solves this problem by offering personalized diet suggestions based
on user-specific data such as age, weight, and height. By employing a content-based recommendation
engine, the system analyzes nutritional information about food items to generate customized
recommendations that align with individual preferences and dietary restrictions. The solution ensures that
the recommendations are not only tailored but also diverse, providing users with a variety of healthy meal
options. With the use of Docker for deployment, the application can easily be scaled and run in any
environment, offering users a consistent and efficient experience. Additionally, the system can be accessed
through a user-friendly web interface built with Streamlit, making it accessible and intuitive for all users.
METHODOLOGY
•User Input Interface: Users enter basic details—age, height, weight, activity level.
•BMI Calculation Module:
•BMI = weight (kg) / (height in meters²)
•Categories: Underweight, Normal, Overweight, Obese
•Health Status Analysis:
•Classify users into BMI categories.
•Trigger specific recommendation plans based on the category.
•Machine Learning Model for Recommendation: .
•A Content-Based Filtering ML model is used.
•Trained on a dataset containing food items, calories, nutritional value, dietary tags (e.g., keto, protein-rich, low-carb).
•Recommends meals matching user profile + health goals.
•Visualization:
•Uses pie charts to show nutritional breakdown (carbs, protein, fat).
•Suggests food recipes with calorie values and macros.
•Weight Management Plan:
•Based on BMI, a tailored weight gain/loss/maintenance plan is generated.
TECH STACK
🧰 Tech Stack

Frontend
•React.js / Vue.js
•HTML5, CSS3, JavaScript
•Chart.js (for Pie Chart Visualizations)

🧠 ML Model (Backend)
•Python
•scikit-learn, Pandas, NumPy, Matplotlib
•Flask / FastAPI (for API)
•Docker (Containerization)

Database
•Firebase Firestore / MongoDB
•Firebase Authentication

Deployment
•Docker
•Render / Heroku / Vercel / Google Cloud
•GitHub for version control

🧭 Architecture Diagram
Includes:
•User Input ➜ BMI Calculation ➜ Classification ➜ ML Model ➜ Diet & Food Recommendation ➜ Visualization (Pie
Chart)
ARCHITECTURE DIAGRAM
ARCHITECTURE WORKFLOW
LIST OF MODULES
1. User Profile & Input Module
• Handles: Age, weight, height, gender, preferences, allergies, goals.
• Functions:
• BMI Calculation
• Preprocessing for input validation
• Frontend: Streamlit form interface
• Backend: FastAPI handles form data as API calls

2. 🍱 Diet Recommendation Engine Module (Content-Based ML Model)


• Machine Learning logic:
• Based on Nearest Neighbors (cosine similarity)
• Utilizes preprocessed nutrition data
• Functions:
• Automatic diet recommendation
• Personalized based on user metrics
• Library Used: scikit-learn

3. 🍔 Custom Food Recommendation Module


• User specifies: Macronutrients, calorie range, ingredient types, etc.
• Functions:
• Searches food database using filtering
• Uses cosine similarity for closest match
• Frontend: Handled via Streamlit sidebar
• Backend: FastAPI routes that filter recipes and return JSON
LIST OF MODULES
4. 📦 Food/Nutrition Database Management Module
•Handles:
•Food.com dataset (500K recipes, 1.4M reviews)
•Nutrition facts per item
•Data Operations:
•Cleaning using pandas
•Preprocessed features stored for ML
•Backend Access: Food data queried via FastAPI

5. 🧠 Model Training & Inference Module


• Functions:
• Train Nearest Neighbors model using nutrition vectors
• Save model state or reload for prediction
• Cosine similarity computation
• Libraries: NumPy, scikit-learn
• Deployment: Included in Docker container

6. 🌐 API Communication Module


•Built with: FastAPI
•Endpoints:
•/predict_diet for automatic diet
•/recommend_food for custom suggestions
•Functions:
•Handles user data requests
•Connects backend logic to frontend
LIST OF MODULES

7. Frontend Interface Module (Streamlit)


•Pages:
•hello.py - Intro page
•auto_diet.py - Auto diet recommendation
•custom_food.py - Custom nutrition-based recommendation
•Features:
•Interactive UI
•Visualizations with streamlit-echarts
•Sidebar navigation

8. 🐳 Deployment & Containerization Module (Docker + Docker Compose)


•Dockerized Services:
•FastAPI backend
•Streamlit frontend
•Compose Services:
•Defined in docker-compose.yaml
•Allows for easy container spin-up
•Benefits: Portability, Scalability, Consistency
THANK YOU

You might also like