A PROJECT REPORT ON

Smart Recipe Modifier : AI – Powered

Indian Diet Customization

Submitted in partial fulfillment of the requirements

of the degree of
BACHELOR OF ENGINEERING
Computer Science & Engineering
(Artificial Intelligence & Machine Learning)

by

1. Vaishnavi Mule (TEA170)

2. Rishikesh Yadav (TEA171)

3. Arya Rane (TEA173)

4. Saniya Pawar (TEA176)

Guide
Prof. Himali Patil

Department of Computer Science & Engineering

(Artificial Intelligence & Machine Learning)
Lokmanya Tilak College of Engineering
Sector-4, Koparkhairne, Navi Mumbai
(2025-2026)
 CERTIFICATE

This is to certify that the Mini Project entitled “Smart Recipe

Modifier: AI – Powered Indian Diet Customization” is a Bonafide

work of Vaishnavi Mule [AIML 170], Rishikesh Yadav [AIML

171], Arya Rane [AIML 173], Saniya Pawar [AIML 176]

submitted in partial fulfillment of the requirement for the award of

the degree of “Bachelor of Engineering” in “Computer Science &

Engineering (Artificial Intelligence & Machine Learning)”.

Prof. Himali Patil

(Guide Name)

(Dr. Chaitrali Chaudhari)

Head of Department
 Mini Project Approval

This Mini Project entitled “Smart Recipe Modifier: AI – Powered

Indian Diet Customization” by Vaishnavi Yadav [AIML 170],

Rishikesh Yadav [AIML 171], Arya Rane [AIML 173], Saniya Pawar

[AIML 176] is approved for the degree of Bachelor of Engineering in

“Computer Science & Engineering (Artificial Intelligence & Machine

Learning)”

Examiner:

1………………………………………

2………………………………………

(External Examiner name & Sign)

Date:

Place:
 Declaration

I declare that this written submission represents my ideas in my own words

and where others' ideas or words have been included, I have adequately cited
and referenced the original sources. I also declare that I have adhered to all
principals of academic honesty and integrity and have not misrepresented or
fabricated or falsified any idea/ data / fact
/ Source in my submission. I understand that any violation of the above will
be cause for disciplinary action by the Institute and can also evoke penal action
from the sources which have thus not been properly cited or from whom proper
permission has not been taken when needed.

1. Vaishnavi Mule (TEA170)

2. Rishikesh Yadav (TEA171)

3. Arya Rane (TEA173)

4. Saniya Pawar (TEA176)

Date:
 TABLE OF CONTENTS
Abstract................................................................................................................... I
Acknowledgment ...................................................................................................II
Table of contents .................................................................................................. III

Chapter 1. Introduction
1.1 Introduction…………………………………………………………………
1.2 Motivation………………………………………………………………….
1.3 Statement of the Problem ………………………………………………...
Chapter 2. Literature Survey
2.1 Survey Existing System (referenced research paper)
………………………
2.2 Limitations of Existing system or research
gap…………………………….
2.3 Objective
…………………………………………………………………...
2.4 Scope of the Work
………………………………………………………….
Chapter 3. Proposed System
3.1 Details of Hardware &
Software……………………………………………
3.2 Design
details……………………………………………………………….
3.3 Methodology
……………………………………………………………….
Chapter 4. Results Analysis
Chapter 5. Conclusion & Future scope
Chapter 6. References (Books, journals and other online references)
.......................…

Annexure (if applicable): Any paper presentation, research funding,

sponsorship information/ certificate may be included.
 Acknowledgment

I remain immensely obliged to [Link] Patil. for providing us

with the idea of this topic, and for her invaluable support in gathering
resources for us either by way of information or computer also her
guidance and supervision which made this project successful.

I would like to thank Head of CSE(AI&ML)Department [Link]

Chaudhari , Principal, LTCoE Dr. S. K. Shinde .

I am also thankful to faculty and staff of Computer Science &

Engineering (Artificial Intelligence & Machine Learning), Department
and Lokmanya Tilak College of Engineering, Navi Mumbai for their
invaluable support.
I would like to say that it has indeed been a fulfilling experience for
working out this project topic.
 1. Introduction

1.1 Introduction
The Smart Recipe Modifier: AI-Powered Indian Diet Customization project
focuses on automatically adapting traditional Indian recipes to meet the dietary
preferences and health requirements of individuals. Using Artificial Intelligence (AI)
and Machine Learning (ML), the system analyzes user input such as dietary type—
low-carb, high-protein, diabetic-friendly, vegetarian, or vegan—and recommends
suitable ingredient substitutions while preserving the dish’s cultural authenticity and
taste. The Smart Recipe Modifier leverages AI and ML to simplify this task. Users
can input any Indian recipe through a web interface, and the system automatically
generates a modified version that suits their chosen dietary preference while
maintaining flavor and authenticity. The platform uses modern technologies like
React, TypeScript, Tailwind CSS, JavaScript, and MongoDB to ensure seamless,
secure, and efficient functionality.

1.2 Motivation
As awareness of nutrition and health grows, people increasingly look for diet plans
tailored to their specific conditions or fitness goals. However, most recipe applications
either generate new dishes or provide general suggestions that fail to consider cultural
nuances, especially in Indian cuisine.
Our motivation is to create an intelligent system that personalizes recipes based on
user preferences and health goals, making healthy eating effortless without
compromising the traditional essence of food. The project aims to empower users with
quick, AI-driven modifications and promote sustainable, balanced lifestyles.

1.3 Statement of the Problem

With growing awareness of health and nutrition, people are increasingly adopting
specific diets such as low-carb, high-protein, diabetic-friendly, vegetarian, or vegan.
However, modifying traditional Indian recipes to meet these diverse dietary needs is
a complex and time-consuming process, especially for individuals with limited
nutritional knowledge.
Existing recipe applications mainly provide new recipe suggestions rather than
modifying the ones users already have. They also fail to offer accurate ingredient
substitutions suitable for Indian cuisine, leading to a loss of authenticity in taste and
cultural value. There is therefore a need for an intelligent system that can automatically
customize recipes based on dietary requirements while maintaining nutritional balance
and traditional flavor. By using Artificial Intelligence (AI) and Machine Learning (ML),
the system can provide accurate, real-time modifications that make healthy eating simple
and accessible to all users.
 Chapter 2. Literature Survey
2.1 Survey Existing System (referenced research paper)
1. The paper “Generating Smart Recipe Using Artificial Intelligence “ by
Sinchana K N, Jeevan Gowda L S, Nithin M G, Mohith H P, Shashidhara H V,
Department of CSE, Malnad College of Engineering, Hassan, India
International Journal of Innovative Research in Technology (IJIRT), Volume 12,
Issue1(2025) This research presents a multi-modal AI culinary platform that
generates personalized, nutritionally balanced, and culturally sensitive recipes
from text, image, and voice inputs. Using ML, computer vision (YOLOv7), NLP
(BERT, GPT-3.5), and graph neural networks, it adapts recipes to dietary needs,
available ingredients, time limits, and user preferences.
2. The study “ Enhancing Recipe Generation Using AI and ML” by Akshara
Pareek, Aastha Kanwar Gahlot, Neha Khatri, Mohnish Sachdeva. International
Journal of Novel Research and Development (IJNRD) (2024) This work
explores various AI/ML approaches—such as NLP, recommender systems,
and GANs—for automating recipe creation. It addresses challenges like
modeling ingredient combinations, cultural sensitivity, user feedback
integration, and generating novel yet context-aware recipes.
3. “Recipe-Fusion: Multimodal Food Recipe Recommendation System “
Karthik Anilkumar Nair, Sanjog Gajanan Chavhan, Tejas Nilkanth Pawar,
Anmol Jagdish Reddy, and K.S. Charumathi. Journal of Artificial Intelligence
Research & Advances (JoAIRA), Volume 11, Issue 03(2024) The Recipe-
Fusion system is a multimodal recipe recommender that uses both visual and
textual inputs to suggest recipes. It employs a CNN to assess image similarity
for visual-based recommendations and an RNN to predict recipes from user-
entered text. The system filters suggestions by diet type (e.g., vegetarian/non-
vegetarian) and course type, aiming to simplify recipe discovery through a
fusion of both machine learning methods
4. Finally“AI-Assisted Plant-Based Recipe Generation “ to Support Sustainable
and Healthy Diets by Anna Thomas, PhD, Dan Jurafsky, PhD, Kristina
Gligorić, PhD, Stanford University (2024) This study investigates how large
language models (LLMs), trained on extensive food-related datasets (including
recipes, reviews, and research articles), can assist in the shift to plant-based
diets. Specifically, the researchers explore whether LLMs can predict human
preferences between pairs of recipes.
 2.2 Limitations of Existing system or research gap
The current systems exhibit several major limitations. Most recipe applications are
designed to generate completely new dishes instead of intelligently modifying the
user’s existing recipe. Ingredient substitutions are not accurate or culturally suitable
for Indian cuisine, leading to loss of authenticity in flavor. Additionally, these
systems provide little to no personalization for specific diet plans like diabetic-
friendly, vegan, or high-protein options.
Manual recipe modification also remains a time-consuming and confusing task,
especially for beginners who may not have adequate knowledge of nutrition or
substitution methods. Moreover, the combination of AI with user preferences for
real-time recipe transformation is largely missing from current
[Link], there is a clear research gap in developing a system that
integrates AI-based customization with user input to generate instant, accurate, and
culturally relevant recipe modifications.

2.3 Objective
The primary objective of the Smart Recipe Modifier: AI-Powered Indian Diet
Customization project is to create an intelligent and user-friendly system that can
automatically modify traditional Indian recipes according to specific dietary
preferences. The project aims to assist individuals in maintaining a balanced diet by
adapting recipes into forms such as low-carb, high-protein, diabetic-friendly,
vegetarian, or vegan, while preserving their cultural authenticity. This ensures that
users do not have to compromise on flavor or traditional essence while pursuing
healthy eating habits.

Another key objective is to reduce manual effort in recipe customization. Many

people find it difficult and time-consuming to manually search for substitute
ingredients or adjust nutritional values in recipes. The proposed system will analyze
the original recipe and automatically recommend ingredient replacements that align
with the user’s dietary goals. By doing this, the project aims to save time, minimize
confusion for beginners, and simplify the entire process of diet-based cooking.

The system also focuses on integrating Artificial Intelligence (AI) and Machine
Learning (ML) to make recipe customization more efficient and accurate. AI
algorithms will be trained to understand different diet categories, identify
ingredients that do not fit those categories, and suggest alternatives that retain
similar taste, texture, and nutritional value. Another major objective is to promote
health awareness and sustainability. The system encourages users to make
mindful food choices that support both their health and the environment.

The system is designed to be compatible with modern web technologies such as

React, TypeScript, Tailwind CSS, and MongoDB, ensuring an intuitive interface
and smooth performance. In the future, the system can be expanded to include
multilingual recipe support, integration with nutrition tracking apps, and a
mobile version to make personalized healthy cooking accessible to everyone.
 2.4 Objective

The primary objective of the Smart Recipe Modifier: AI-Powered Indian Diet
Customization project is to create an intelligent and user-friendly system that can
automatically modify traditional Indian recipes according to specific dietary preferences. The
project aims to assist individuals in maintaining a balanced diet by adapting recipes into
forms such as low-carb, high-protein, diabetic-friendly, vegetarian, or vegan, while
preserving their cultural authenticity. This ensures that users do not have to compromise on
flavor or traditional essence while pursuing healthy eating habits.

Another key objective is to reduce manual effort in recipe customization. Many people
find it difficult and time-consuming to manually search for substitute ingredients or adjust
nutritional values in recipes. The proposed system will analyze the original recipe and
automatically recommend ingredient replacements that align with the user’s dietary goals.
By doing this, the project aims to save time, minimize confusion for beginners, and
simplify the entire process of diet-based cooking.

The system also focuses on integrating Artificial Intelligence (AI) and Machine
Learning (ML) to make recipe customization more efficient and accurate. AI algorithms
will be trained to understand different diet categories, identify ingredients that do not fit
those categories, and suggest alternatives that retain similar taste, texture, and nutritional
value. This makes the platform not just a recipe modifier, but a smart cooking assistant
that continuously improves its suggestions based on user interactions and feedback.

Another major objective is to promote health awareness and sustainability. The system
encourages users to make mindful food choices that support both their health and the
environment. By recommending locally available and healthier substitutes, it helps in
reducing dependence on processed or imported ingredients, thus promoting sustainable
eating habits. The project also aims to contribute to reducing food waste by providing
alternative uses for leftover ingredients within recipe suggestions.

Finally, the project aims to build a scalable, accessible, and secure web-based platform
where users can easily input recipes and receive personalized results in real time. The
system is designed to be compatible with modern web technologies such as React,
TypeScript, Tailwind CSS, and MongoDB, ensuring an intuitive interface and smooth
performance. In the future, the system can be expanded to include multilingual recipe
support, integration with nutrition tracking apps, and a mobile version to make
personalized healthy cooking accessible to everyone.
Chapter 3. Proposed System

3.1 Details of Hardware & Software

Hardware Requirements
 Processor: Intel i3 or higher
 RAM: 4 GB (8 GB recommended)
 Storage: 256 GB SSD or higher
 Operating System: Windows / macOS / Linux
 Internet connection for online processing and data access
Software Requirements
 Frontend: [Link], TypeScript, Tailwind CSS
 Backend Logic: JavaScript ([Link])
  Database: MongoDB
 Build Tool: Vite
 Development Environment: Visual Studio Code
 Deployment Platforms: Netlify / GitHub Pages

3.2 Design details:

 Requirement Analysis:
Collected user requirements such as recipe input, diet preference selection, and
instant modified output display for personalized use.

 System Design:
Designed architecture using modular components for input processing, AI-based
ingredient mapping, and result generation.

 User Interface Design:

Created an intuitive, responsive interface using [Link] and Tailwind CSS,
ensuring easy navigation and accessibility.

 Logic Development:
Implemented recipe modification logic in [Link], managing ingredient
substitution using mappings from [Link].

 Testing & Validation:

Conducted functionality and performance tests to ensure accurate substitutions,
fast response, and consistent UI behavior across devices.

 Deployment:
Deployed final build using Vite and hosted on Netlify/GitHub Pages for smooth
online access and real-time functionality.
 3.3 Methodology
The project follows a systematic software development life cycle (SDLC)
approach to ensure accuracy, usability, and efficiency. Each phase focuses on
building a reliable and intelligent recipe modification system using AI and ML
techniques.
 Step 1: Requirement Analysis
Identified user needs such as recipe input, dietary preference selection, and quick
AI-based recipe modification.

 Step 2: Planning
Defined workflow, selected technologies ([Link], TypeScript, Tailwind CSS,
MongoDB), and designed substitution logic.

 Step 3: System Design

Divided system into modules for input, AI-based processing, and modified
output generation using structured data flow.

 Step 4: Development
Implemented core logic in [Link], ingredient mapping in
[Link], and integrated all React components.

 Step 5: Testing
Conducted functional and UI testing to ensure accurate substitutions, speed, and
responsiveness across devices.

 Step 6: Deployment & Maintenance

Deployed final build using Vite and hosted on Netlify/GitHub Pages; periodic
updates enhance diet types and UI design.

Implementation: The implementation phase focuses on converting the design into a

functional system using modern web technologies. Each component is developed, tested,
and integrated to ensure smooth performance, accurate AI-driven recipe modification, and
a user-friendly interface.

 Development Environment Setup and Deployment:

Created the project using npm create vite@latest, configured React,
TypeScript, and Tailwind [Link] optimized build created using npm run
build and hosted on Netlify/GitHub Pages for online accessibility and smooth
user experience.

 Frontend Implementation:
Built UI components for recipe input, dietary selection, and result display using
[Link] and Tailwind CSS for responsiveness.
  Backend Logic:
Implemented the main recipe modification logic in [Link],
which processes user input and substitutes ingredients based on
[Link] mapping.

 Integration and Testing Phase:

Combined all components in [Link] for seamless interaction between UI and
logic [Link] local testing using npm run dev to verify functionality,
substitution accuracy,and responsiveness.

Chapter 4. Results Analysis

The The Smart Recipe Modifier project was analyzed at each development stage to
ensure accuracy, usability, and performance. The system was designed to
intelligently modify recipes using AI, ML, and NLP, providing accurate ingredient
substitutions for various diets such as diabetic-friendly, vegan, and high-protein.

During the Requirement Analysis phase, user needs were carefully studied to design
features like the recipe input box, diet preference selector, and real-time modified output
display. This helped define a smooth workflow and user-centric interface for better
accessibility.

In the System Design and Development phase, component structures were defined using
[Link], TypeScript, and Tailwind CSS, while the substitution logic was implemented
in [Link]. The ingredient mapping was maintained in [Link], ensuring
each dietary preference generated accurate substitutions without losing cultural
authenticity.

Extensive testing and validation were performed using multiple recipes and diet
combinations. The system successfully replaced non-compliant ingredients while
maintaining the nutritional value and original flavor of Indian dishes. The use of NLP-
based tokenization improved the accuracy of ingredient recognition and replacement.

The system achieved high efficiency and user satisfaction during testing. Its interface
responded quickly, providing modifications in seconds with minimal errors. Overall, the
project effectively demonstrated how AI and ML can simplify healthy eating by bridging
traditional cooking with modern dietary needs.

Overall, the analysis confirms that the Smart Recipe Modifier effectively bridges the gap
between traditional cooking and modern nutrition. It promotes healthier eating habits,
supports personalized diets, and maintains the cultural integrity of Indian cuisine through
advanced AI and ML technologies. The successful results indicate the system’s potential
as a foundation for future smart cooking and dietary recommendation
 Chapter 5. Conclusion & Future scope
Conclusion:

The Smart Recipe Modifier: AI-Powered Indian Diet Customization project

successfully demonstrates the integration of Artificial Intelligence (AI) and Machine
Learning (ML) in enhancing daily cooking and nutrition practices. The system provides an
innovative solution to a common problem — adapting traditional Indian recipes to meet
specific dietary requirements such as diabetic-friendly, vegan, low-carb, or high-protein
diets. By leveraging AI algorithms and Natural Language Processing (NLP), the system
intelligently identifies non-compliant ingredients and replaces them with healthier,
culturally suitable alternatives. This not only helps users maintain dietary discipline but
also preserves the taste and authenticity of Indian food.
Through systematic development and implementation, the project demonstrated that AI
can be effectively used for real-time personalization in food systems. The integration of
[Link], TypeScript, and Tailwind CSS in the front end provided a seamless and
responsive interface, while the backend logic handled intelligent substitutions using
structured ingredient mapping. The system was tested with multiple recipes under varied
diet preferences and consistently produced accurate, balanced results. The use of NLP
improved text recognition and tokenization accuracy, allowing precise identification of
ingredients and their replacements. Overall, the system exhibited high performance,
accuracy, and scalability — validating its reliability as an AI-powered assistant for
personalized meal customization.

The success of this project lies not only in meeting its technical objectives but also in
promoting a health-conscious and sustainable lifestyle. It empowers users to make
informed dietary choices while preserving traditional Indian cooking practices. The Smart
Recipe Modifier bridges the gap between technology, culture, and health, serving as a
stepping stone for future advancements in AI-driven nutrition systems. Its modular design
allows easy enhancement, making it adaptable to emerging food trends and global cuisines.
The system’s strong foundation opens doors for future integration with health monitoring
devices, multilingual recipe databases, and mobile platforms, transforming it into a
comprehensive digital culinary companion. In conclusion, this project stands as a testament
to how intelligent technology can enrich everyday life by merging tradition with innovation
for a smarter, healthier future.

In conclusion, the Smart Recipe Modifier stands as a step forward in combining

technology, culture, and health awareness into one intelligent platform. Its success opens
new opportunities for expanding the technology into broader applications such as nutrition
tracking, multilingual recipe support, and mobile-based platforms. The project not only
fulfills its initial objectives but also establishes a foundation for future innovations in smart
cooking and AI-powered dietary solutions.
 Future Scope:
The Smart Recipe Modifier project has immense potential for future development and
enhancement. With rapid advancements in Artificial Intelligence and the growing
awareness of health-conscious living, the system can be expanded into a more
comprehensive and intelligent platform that assists users in every aspect of healthy eating.
The following points highlight the key areas where the system can be improved and
extended in the future:

1. Voice and Image Input:

Future versions can allow users to upload recipe images or give voice commands for
instant AI-based modification, making the system more interactive and user-friendly.

2. Nutrition Analysis and Meal Planning:

Integrating calorie and nutrition tracking will help users monitor their daily intake and
achieve personalized health and fitness goals.

3. Regional Language Support:

Adding multiple Indian languages such as Hindi, Marathi, and Tamil will make the
application accessible to a wider audience across diverse regions.

4. Global Cuisine Expansion:

The system can be expanded to support recipe modification for international cuisines,
allowing users to customize global dishes based on their dietary preferences.

5. Integration with Fitness Apps:

Connecting with health and fitness tracking applications will create a complete healthy-
living ecosystem linking diet, activity, and wellness.

6. Smart Kitchen & IoT Integration:

Future versions can integrate with smart kitchen appliances and IoT devices to suggest
recipes based on available ingredients and automate certain cooking processes.

7. Mobile Application Development:

Developing a mobile version will provide users with easy accessibility and real-time
recipe modification anytime, anywhere.

8. Machine Learning Enhancements:

With continuous learning, ML models can be refined to deliver more accurate ingredient
substitutions and smarter personalization over time.
 Chapter 6. References (Books, journals
and other online
references)
Research papers:
1. AI-Based Food Recommendation System for Personalized Diet Planning –
IJFANS (2023)
[Link]

2. Artificial Intelligence in Nutrition and Recipe Management – IJRIT (2022)

[Link]

3. AI-Powered Recipe Customization and Food Personalization using NLP and

ML – IEEE Xplore (2024)
[Link]

4. FoodAI: Machine Learning for Nutrition Management and Ingredient

Substitution – ResearchGate (2023)
[Link]

Websites:

1. AllRecipes – Global Recipe Database & Ingredient Substitution Guide

[Link]

2. EatThisMuch – Automated Nutrition & Meal Planning Tool

[Link]

3. MyFitnessPal – Calorie Counter & Nutrition Tracker

[Link]

GitHub Repositories:

1. AI Recipe Recommender using NLP

[Link]

2. Food Recommendation System using Machine Learning

[Link]