Flavour Fusion : A Recipe Recommendation based

on Ingredient
Mr Ansari Mohd Aamir Mr Ehtesham Ahmed Masood Ahmed Mr Patel Sufiyan
Dept of CSE(AIML) Dept of CSE(AIML) Dept of CSE(AIML)
M.H Saboo Siddik College of M.H Saboo Siddik College of M.H Saboo Siddik College of
Engineering Engineering Engineering
[email protected] [email protected] Mumbai , India
[email protected]
Ms Sumaiya Rauf Memon Mrs Chaitali Mahajan
Dept of CSE(AIML) Dept of CSE(AIML)
M.H Saboo Siddik College of M.H Saboo Siddik College of
Engineering Engineering
Mumbai , India Mumbai , India
[email protected]

Abstract websites with cooking recipes, some of which are "easy" to

prepare. Nevertheless, those recipes are not rated as "easy"
Cooking has become more of a challenge in the current world based on the requirements of the user. To create a
since both men and women work, and the younger generation recommendation system, we must first design a mechanism
can't devote enough time to learning how to cook since they are
to suggest "easy" cooking recipes by examining the
too busy studying. One of the most important things we can do in
our daily lives is to know what to prepare and how to make it.
substance of the recipes and taking the user's circumstances
However, how can everyone, whether they a college student, a into account. [5]. When there is an abundance of
businesswoman or man, or a hosteller, cook easily? To address information, a recommendation system narrows it down to
this problem, we have created a website that, by looking up just the most interesting stuff. This acts as an information
ingredient names, provides recipes for a variety of foods prepared filter. A vast variety of information on goods, services, and
all over the world. With a wide selection of recipes from media material can be filtered by the recommender system.
throughout the globe, our cutting-edge website makes cooking The recommender system's objective is to provide a group
easier for those who lead busy lives. By utilizing dynamic API of users with relevant recommendations for goods or
connectivity, we guarantee accuracy and real-time changes. The
services that they would find interesting. From the
website specializes on Indian food and offers a wide variety of
both traditional and modern dishes. With recipes divided into standpoint of e-commerce, a recommender system is a tool
breakfast, lunch & supper, and desserts, users may easily choose that assists users in searching through knowledge records
meals that suit their unique taste buds and time restrictions. linked to their interests and preferences.[6] Because there
are so many options for cooking, selecting the correct extra
components and recipe samples is a crucial yet difficult task
Keywords—Ingredients ,Recipe ,Website, Recommendation in the food industry.[7] Several apps and websites are
available in the culinary business that allow users to search
for recipes using keywords like ingredient names, cuisine
types, or cultural references. While these programs cater to
I. INTRODUCTION
the interests and demands of their users, they do not specify
People today are trying to include as much healthy food as limitations, like the upper limit on the quantity of materials.
possible in their meals since they are becoming more and Users then either forget to add components or substitute
more health concerned[1] Humans require food, drink, air, identical ones. Ingredient recognition can be used to
and shelter to survive. Food is essential to our survival. determine which components should be avoided by
Therefore, a person with cooking knowledge has an consumers to help them.[8] The Indian Recipe
advantage over others. People are investing more in their Recommendation System is a system that suggests new,
health and fitness thanks to technological improvements, but untested recipes to users based on their prior preferences for
eating food as medicine is necessary to achieve this aim.[2] certain foods. The components in the recipes that the user
Cooking recipe websites like cooks.com and BBC food has already expressed a preference for serve as the
search have gained popularity recently. These websites are foundation for recommendations.[9] A domain-specific
used by some cooks to find recipe information. A user can subclass of recommender systems called "food
view the cooking recipe websites at home or at a grocery recommender systems" was created to help consumers find
shop because they are available on both PCs and mobile recipes or meal plans that suit their dietary requirements.
devices.[3] A well-liked platform for exchanging recipes Food recommenders, which often use a count to show the
and culinary techniques from various cuisines is recipe collective opinion of users, unavoidably face the following
websites. For those who wish to taste new recipes they issues, which are difficult to resolve by using current
would never try and are searching for their favorite dishes, techniques.[10] We draw inspiration from the current
they open a new door.[4] Nowadays, there are a ton of system, in which the user is limited to thinking about dishes
that he has been introduced to through cookbooks, cooking learning, Desai et al. (2019, 2021) contributed to the subject
programs, moms' or grandmothers' recipe books, or based by developing a recipe recommendation system based on
on his own culinary or dining experiences.[11] Because of ingredients [5, 11]. Through the application of conventional
the ingredients, the flavor, or a friend's tip, a user can machine learning techniques, their work investigates how to
choose to attempt a new dish. Consequently, careful recipe integrate statistical models for efficient recommendation.
suggestion ought to include each of these aspects.[12] The "Recommendation System of Indian Cuisine Recipes
To make the process of discovering new meals enjoyable Using Content-Based Approach" was thoroughly reviewed
and effortless, this project aims to create an interactive and by Dalal et al. (2020) [6]. This study, which focused on
intuitive culinary companion that changes with the user. Indian food, demonstrated the value of content-based
strategies in adjusting suggestions to cultural tastes.
II. RELATED WORK "Recipe Bowl" by Gim et al. (2021) [7] is a clear example
In this section, we introduce related works in terms of flavor of how transformer designs are used into recipe
fusion A Recipe Recommendation based on ingredients. recommendation systems. To demonstrate the adaptability
and efficiency of transformer-based models in capturing
intricate interactions within recipe data, this culinary
recommender uses set transformers for both ingredients and
recipes.
A recommendation system called "RecipeIS," which is
based on the identification of food ingredients, was
introduced by Rodrigues et al. in 2023 [8]. The work they
have done combines sophisticated recommendation
algorithms with visual recognition, adding to the increasing
amount of research that highlights the value of visual
information in recipe recommendation.
The "Recommendation of Indian Cuisine Recipes based on
Ingredients" [9] by Nilesh et al. (2019) delves more into the
topic of cultural distinctiveness. Recognising the distinct
taste characteristics and ingredient choices of Indian food,
Recipe recommendation has significantly grown thanks to the writers concentrated on creating recommendations that
advances in computer vision, deep learning, and machine were especially suited for it.
learning. This review of the literature examines important Chen et al. (2020) explored "Eating Healthier: Exploring
studies that advance the creation of intelligent recipe Nutrition Information for Healthier Recipe
suggestion systems, with an emphasis on ingredient-based Recommendation" [10] regarding recipe recommendations
methods. that are health conscious. With the potential to encourage
Morol et al. (2022) conducted a study titled "Food Recipe healthier eating habits through algorithmic suggestions, this
Recommendation based on Ingredients Detection using study highlighted the inclusion of nutritional information in
Deep Learning" [1], which marked the beginning of the the recommendation process.
investigation into ingredient-based recipe recommendation. Tian et al.'s (2022) "Recipe Recommendation with
To identify and detect chemicals, the authors used deep Hierarchical Graph Attention Network" [12] brought graph-
learning techniques, which set the stage for later based techniques to light. The authors added to the
developments in the field. developing field of recommendation systems by
According to Jadhav et al. (2023)"Vision-Based Intelligent demonstrating the power of graph-based models to capture
Recipe Recommendation System" [2], vision-based systems complex linkages in recipe data using a hierarchical graph
are becoming more and more popular. This method attention network.
improves ingredient recognition accuracy by integrating As a result, the literature review highlights the variety of
computer vision technology, which results in more accurate methods used in the field of ingredient-based recipe
recipe recommendations. selection. Each work adds important insights to the
A "Cooking Recipe Recommendation System with Visual overarching objective of enriching the culinary experience
Recognition of Food Ingredients" was developed by Yanai through intelligent and personalized recipe
et al. (2014), who provided an early example of visual recommendations, ranging from deep learning and computer
recognition in recipe recommendation [3]. This vision to machine learning and cultural concerns.
groundbreaking study highlighted how visual cues might
improve the recommendation process.
Nadee and Unankard (2021) offered a "Alternative- III. METHODOLOGY
Ingredient Recommendation based on Correlation Weight
for Thai Recipes" [4] in response to the uniqueness of
regional cuisines. The authors demonstrated the significance First, a thorough requirement analysis is conducted to
of cultural context in ingredient-based systems by identify user demands as part of the recipe suggestion
customizing recommendations for Thai dishes with the use project's process. Using HTML, CSS, and JavaScript, the
of a correlation-weight-based method. design phase focuses on producing an easy-to-use interface.
Recipe recommendations have benefited greatly from the Reactive design and user interaction management are then
application of machine learning techniques. Using machine part of the front-end implementation process. API calls for
ingredient information and external service calls are Flow Chart
managed by the backend, which is designed to handle
server-side operations.[13]The recipe specifics are presented
in an easy-to-use manner, and user inputs are carefully
managed to provide a seamless transition between step
types.[14] Feedback from users is gathered for iterative
improvements after thorough testing and debugging stages.
To make upkeep and future improvements easier, the
procedure is recorded.

IV. RESULT
1) Home page

Here It shows the home screen where we have a Button of

Get Started and Some Information about the project like
Contact Us , About etc.

2) Search Option

Block Diagram

The recipe recommendation project has an interface that is

easy to use because of the usage of HTML, CSS, and
JavaScript. Ingredient-based and choice-based search
options are available on the homepage. The user is prompted Figure shows that we have given two options (1)Search
for ingredient names while using an ingredient-based search, With Ingredients and (2) Select By Your Choice , Here user
which uses an API to retrieve information and cooking can select the required option.
instructions. Choosing choice-based search brings up
screens with recipe categories for breakfast, lunch, supper, 3) Search with Ingredients
and dessert. If you choose a particular category, such as
breakfast, you will see a variety of recipes with thorough
directions and ingredient lists.[15] To improve the user
experience, there are also embedded links to YouTube
videos. In accordance with user preferences, the project
seeks to offer a smooth and interactive platform for users
looking for a variety of recipes.

API: https://www.themealdb.com/api.php
Here the user can enter the ingredients to let us see if we
enter chicken what will be displayed.

After Searching Chicken, we have got Some Recipe from

where user can choose from the given options

The above figure shows some Indian Breakfast Recipe

where we can view recipe.
After Clicking the view Recepie we have all the list of
Ingredients and Steps and here we also provided the Video
Links of the Recipe.

After Clicking View Recipe we have List of Ingredients and

steps to make the recipe as shows in above picture

4) Search By Choice

And If We choose Option two from search option it shows

the Type of Food we need like breakfast, Lunch & Dinner,
and sweets.

V. CONCLUSION
In Conclusion to this , we have created a web page using the
HTML,CSS, and JavaScript-created recipe recommendation
project captures a user-centered method of culinary
exploration. The project's two search options, one for
ingredients and the other for options, serve a wide range of
users looking for customized recipes. Users receive exact
ingredient lists and cooking instructions through the
ingredient-based search's dynamic interaction with APIs.
With categories for breakfast, lunch, dinner, and desserts,
the choice-based search offers a simplified browsing
experience.
By including links to YouTube videos, the project becomes
more useful by giving users visual assistance for a more
enjoyable cooking experience. User engagement and
accessibility are guaranteed by the fluid and responsive
front-end design. The project's adaptability makes it simple
for users to switch between screens, promoting an engaging
and entertaining recipe discovery process.
The project is excellent in offering a variety of breakfast 12) Tian, Y., Zhang, C., Metoyer, R. and Chawla, N.V.,
recipes, with a focus on both practicality and a visually 2022. Recipe recommendation with hierarchical graph
appealing user experience. Operating at the nexus of attention network. Frontiers in big Data, 4, p.778417.
creativity and functionality, it caters to people of every 13) Vivek, M.B., Manju, N. and Vijay, M.B., 2018. Machine
culinary ability. Subsequent improvements such as learning based food recipe recommendation system. In
community features and user profiles are intended to Proceedings of International Conference on Cognition
increase its influence in the digital cooking market. and Recognition: ICCR 2016 (pp. 11-19). Springer
Singapore.
VI. REFERENCES
14) Chavan, P., Thoms, B. and Isaacs, J., 2021. A
recommender system for healthy food choices: building
1) Morol, Md Kishor, et al. "Food recipe recommendation
a hybrid model for recipe recommendations using big
based on ingredients detection using deep
data sets.
learning." Proceedings of the 2nd International
Conference on Computing Advancements. 2022. 15) Swain, M., Manyatha, A.R., Dinesh, A.S., Sampatrao,
G.S. and Soni, M., 2023, February. Ingredients to
2) Jadhav, T., Khatik, V., Navghane, A., Rajpurohit, M.,
Recipe: A YOLO-based Object Detector and
Avghade, K., & Kalshetti, R. Vision Based Intelligent
Recommendation System via Clustering Approach. In
Recipe Recommendation System.
2023 Third International Conference on Artificial
3) Yanai, K., Maruyama, T., & Kawano, Y. (2014). A Intelligence and Smart Energy (ICAIS) (pp. 1397-1404).
Cooking Recipe Recommendation System with Visual IEEE.
Recognition of Food Ingredients. International Journal
of Interactive Mobile Technologies, 8(2).
4) Nadee, Wanvimol, and Sayan Unankard. "Alternative-
ingredient recommendation based on correlation weight
for thai recipes." 2021 Joint International Conference on
Digital Arts, Media and Technology with ECTI Northern
Section Conference on Electrical, Electronics, Computer
and Telecommunication Engineering. IEEE, 2021.
5) Desai, S., et al. "Recipe Recommendation Based on
Ingredients using Machine Learning." International
Journal of Advanced Research in Computer and
Communication Engineering 8.3 (2019).
6) Dalal, Jignasha, et al. "A Review: Recommendation
System of Indian Cuisine Recipes Using Content-Based
Approach." Proceedings of the 3rd International
Conference on Advances in Science & Technology
(ICAST). 2020.
7) Gim, Mogan, et al. "Recipe bowl: A cooking
recommender for ingredients and recipes using set
transformer." IEEE Access 9 (2021): 143623-143633.
8) Rodrigues, Miguel Simões, Filipe Fidalgo, and Ângela
Oliveira. "RecipeIS—Recipe Recommendation System
Based on Recognition of Food Ingredients." Applied
Sciences 13.13 (2023): 7880.
9) Nilesh, Nilesh, et al. "Recommendation of Indian cuisine
recipes based on ingredients." 2019 IEEE 35th
international conference on data engineering workshops
(ICDEW). IEEE, 2019.
10) Chen, M., Jia, X., Gorbonos, E., Hoang, C.T., Yu, X. and
Liu, Y., 2020. Eating healthier: Exploring nutrition
information for healthier recipe recommendation.
Information Processing & Management, 57(6),
p.102051.
11) Desai, S., Patil, M.P., Shinde, M.P., Sayyed, M.A. and
Bhosale, R., 2019. Recipe Recommendation Based on
Ingredients using Machine Learning. International
Journal of Advanced Research in Computer and
Communication Engineering, 8(3).