PERSONALIZED DIET RECOMMENDATION SYSTEM USING ARTIFICIAL NEURAL NETWORKS PDF Free Download
1 / 6/6
100%
e-ISSN: 2582-5208
International Research Journal of Modernization in Engineering Technology and Science
( Peer-Reviewed, Open Access, Fully Refereed International Journal )
Volume:07/Issue:07/July-2025 Impact Factor- 8.187 www.irjmets.com
www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science
[2140]
PERSONALIZED DIET RECOMMENDATION SYSTEM USING ARTIFICIAL
NEURAL NETWORKS
M. Dharani*1, Mr. P. Anbumani*2
*1Master Of Computer Applications, Krishnasamy College Of Engineering &Technology,
Cuddalore, India.
*2MCA., M.Phil., NET, Associate Professor, Master Of Computer Applications, Krishnasamy College Of
Engineering & Technology, Cuddalore, India.
DOI: https://www.doi.org/10.56726/IRJMETS81270
ABSTRACT
This project presents a Personalized Diet Recommendation System utilizing Artificial Neural Networks (ANNs)
to provide tailored nutritional guidance based on individual health parameters such as age, gender, BMI,
activity level, and dietary preferences. The system leverages machine learning to analyze user inputs and
predict optimal meal plans that align with health goals like weight loss, muscle gain, or disease management. By
continuously learning from user feedback, the ANN enhances recommendation accuracy over time. This
intelligent approach aims to promote healthier lifestyles through automated, data-driven diet planning.
Keywords: Personalized Diet, Artificial Neural Networks, Machine Learning, Nutrition, BMI, Health Goals,
Dietary Preferences, Recommendation System, Meal Planning, Data-Driven.
I. INTRODUCTION
In recent years, the importance of personalized nutrition has grown significantly due to the increasing
awareness of how diet affects overall health and well-being. A one-size-fits-all approach to dieting often fails to
accommodate individual differences in metabolism, health conditions, and lifestyle choices. Therefore, there is
a growing demand for intelligent systems that can provide tailored dietary recommendations based on
personal characteristics. Advances in artificial intelligence, particularly in Artificial Neural Networks (ANNs),
have opened up new possibilities for developing such personalized systems. Artificial Neural Networks are
computational models inspired by the human brain, capable of learning complex patterns and relationships
within data. When applied to diet recommendation systems, ANNs can analyze a wide array of personal health
data and generate insights that guide meal planning in a personalized and efficient manner. The use of ANNs
allows for the modeling of non-linear relationships among various factors such as age, weight, activity level,
dietary restrictions, and health goals, which are crucial in creating effective diet plans. Traditional diet planning
methods are often rigid and rely heavily on generalized guidelines that do not consider individual variability.
These methods can be time-consuming and may require the intervention of a nutritionist or dietitian. In
contrast, a system powered by ANNs can process vast datasets and provide instant, user-specific
recommendations, thus making diet planning more accessible and efficient. This system not only saves time but
also increases the likelihood of users adhering to a suitable diet, ultimately leading to better health outcomes.
The personalization of diets involves gathering relevant user data, including physical metrics (like height,
weight, and BMI), dietary preferences (such as vegetarian or vegan), health conditions (like diabetes or
hypertension), and lifestyle factors (such as physical activity and sleep patterns). The ANN is trained on a large
dataset that includes these variables along with successful diet plans and their outcomes. Through training, the
model learns the correlations between user profiles and effective dietary patterns, enabling it to make accurate
future predictions. Once trained, the neural network can predict the most appropriate diet plan for a new user
based on their inputs. The recommendations generated are not static but can be adjusted over time as the user
updates their health metrics or goals. This dynamic adaptation is one of the key advantages of using ANNs in
this domain. It allows the system to evolve with the user’s changing needs and preferences, ensuring continued
relevance and effectiveness. Another important aspect of the system is feedback integration. Users can report
their satisfaction with the recommended diet, track their progress, and highlight any issues such as allergies or
intolerances. This feedback can be used to further train and refine the ANN, making the system smarter and
more personalized with every interaction. In this way, the recommendation engine becomes a continuous
e-ISSN: 2582-5208
International Research Journal of Modernization in Engineering Technology and Science
( Peer-Reviewed, Open Access, Fully Refereed International Journal )
Volume:07/Issue:07/July-2025 Impact Factor- 8.187 www.irjmets.com
www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science
[2141]
learning system that improves with more data and user engagement. Moreover, the use of ANNs in this context
aligns well with current trends in digital health and mobile health technologies. With the increasing use of
smartphones and wearable devices, users can easily track their health data and receive real-time dietary
suggestions through a user-friendly application. The integration of ANN-based recommendation engines with
these platforms can revolutionize how individuals manage their nutrition and health. From a technical
perspective, developing an effective ANN-based diet recommendation system involves several stages, including
data preprocessing, model selection, training, validation, and deployment. Various neural network
architectures, such as feedforward networks or recurrent networks, may be tested to determine which
provides the best predictive performance. The success of the system heavily depends on the quality and
diversity of the training data, as well as the design of the user interface to ensure ease of use and engagement.
This kind of system also holds potential in addressing public health issues such as obesity, malnutrition, and
chronic disease management. By providing affordable and accessible personalized nutrition advice, it can help
reduce the burden on healthcare systems and promote preventive care. Additionally, such systems can
empower users to take greater control over their dietary habits and health decisions, leading to more informed
and health-conscious societies. In conclusion, a Personalized Diet Recommendation System using Artificial
Neural Networks represents a promising advancement in health technology. By leveraging the learning
capabilities of ANNs, the system can deliver accurate, personalized, and adaptable diet plans that cater to the
unique needs of each user. As the technology continues to evolve, it has the potential to become an essential
tool in promoting healthier lifestyles and improving nutritional outcomes for individuals around the world.
II. RELATED WORKS
[1] A Personalized Diet Recommendation System Based on Health Monitoring Data
This paper proposes a personalized diet recommendation system that uses health monitoring data such as
heart rate, blood pressure, and BMI to provide dietary suggestions. By leveraging machine learning algorithms,
especially decision trees and clustering, the system dynamically adjusts the user's food plan. The results
indicate that integrating real-time health data improves the relevance and accuracy of nutritional
recommendations.
[2] Food Recommendation System for Diabetic Patients Using Machine Learning
The paper presents a food recommendation system designed for diabetic patients by employing supervised
machine learning techniques. The system considers age, gender, weight, dietary habits, and diabetic conditions.
Using a Naïve Bayes classifier, the system predicts suitable food options with high accuracy. The study
emphasizes the importance of health parameters in delivering tailored diet suggestions.
[3] AI-Based Personalized Diet Planner Using Nutritional Ontology and ANN
This research uses artificial neural networks (ANN) along with nutritional ontology to create a personalized
diet planner. Inputs such as age, gender, health conditions, allergies, and exercise routines are taken into
account. The ANN model recommends meals by mapping health goals with dietary constraints. Results show
improved user satisfaction and dietary adherence over traditional plans.
[4] Nutritional Food Recommender System Using Hybrid Filtering Techniques
This study introduces a hybrid filtering approach combining collaborative and content-based filtering for food
recommendation. It integrates user preferences, physical activity levels, and BMI to recommend personalized
meals. The system uses a fuzzy logic model to enhance decision-making under uncertainty. The hybrid model
outperformed conventional single-method recommenders in terms of accuracy and user personalization.
[5] Personalized Meal Recommendation for Healthy Living Using Deep Learning
This paper focuses on deep learning methods, particularly convolutional neural networks (CNNs), to analyze
user profiles and generate healthy meal plans. Key features such as dietary preferences, allergies, nutritional
requirements, and physical stats are used. The system adapts over time with user feedback, showing
improvements in health indicators. It concludes that deep learning enables highly adaptive and scalable
recommendation systems.
Celestine Iwendi et al. (2020) explore the data collection potential of their system, focusing on machine and
deep learning algorithms such as Naive Bayes, Logistic Regression, Multilayer Perceptron (MLP), Gated
e-ISSN: 2582-5208
International Research Journal of Modernization in Engineering Technology and Science
( Peer-Reviewed, Open Access, Fully Refereed International Journal )
Volume:07/Issue:07/July-2025 Impact Factor- 8.187 www.irjmets.com
www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science
[2142]
Recurrent Units (GRU), Recurrent Neural Networks (RNN), and Long Short-Term Memory (LSTM). They
collected information on 30 people with 13 different illnesses and 1000 items from the internet and hospitals
for inclusion in the clinical dataset. The system features eight product area attributes. Before applying deep
learning and machine learning techniques, the characteristics of this Internet of Medical Things (IoMT) data
were examined and encoded [6].
The USDA nutrition dataset will determine the user's suggested diet, incorporating grocery shop information
based on the user's preferred food intake. This database contains nutritional data for every food item, using a
USDA ID as the baseline value for input values per 100 grams. Since these values are crucial for estimating the
suggested diet, BMI data must be provided. This approach is discussed by Butti Gouthami and Malige Gangappa
(2020)
III. PROPOSED SYSTEM
The proposed system is designed to deliver personalized nutritional food recommendations using Artificial
Neural Networks (ANNs). Initially, user data is collected, including demographic details such as age and gender,
along with health and lifestyle-related information. This data undergoes preprocessing to remove noise, handle
missing values, and convert it into a suitable format for analysis. Next, feature extraction techniques are applied
to identify key attributes that influence dietary needs, followed by feature selection to retain only the most
relevant features for accurate predictions. The refined data is then fed into an ANN-based classification model
that predicts the user's age and gender. These predicted attributes are crucial inputs for the recommender
system, which generates personalized diet suggestions tailored to the individual's nutritional requirements. By
integrating machine learning with a recommendation engine, the system ensures that users receive health-
conscious food recommendations based on their unique profile.
Figure 1: System Architecture of proposed system
IV. MODULES
Data Collection
Preprocessing
Features Extraction
Classification
Diet Recomemdation
Data Collection
The recommended diet for the user will be determined using the USDA nutrition information. Every food item's
nutrition information is maintained in the USDA database.
Description: Collect essential personal and health data from the user.
Inputs:
o Age
e-ISSN: 2582-5208
International Research Journal of Modernization in Engineering Technology and Science
( Peer-Reviewed, Open Access, Fully Refereed International Journal )
Volume:07/Issue:07/July-2025 Impact Factor- 8.187 www.irjmets.com
www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science
[2143]
o Gender
o Height
o Weight
o BMI (calculated)
o Daily meal preferences (veg/non-veg/vegan)
o Exercise intensity (none, light, moderate, heavy)
o Weight goal (lose, maintain, gain)
o Food allergies (if any)
Preprocessing
Data pre-processing is an important step in the [data mining] process. The phrase "garbage in, garbage out" is
particularly applicable to data mining and machine learning projects. Data-gathering methods are often loosely
controlled, resulting in out-of-range values, impossible data combinations, missing values, etc. Thus, the
representation and quality of data is first and foremost before running an analysis. If there is much irrelevant
and redundant information present or noisy and unreliable data, then knowledge discovery during the training
phase is more difficult. Data preparation and filtering steps can take considerable amount of processing time. In
this module, eliminate the irrelevant and missing values in uploaded datasets.
Features Extraction
Feature extraction is a general term for methods of constructing combinations of the variables to get around
these problems while still describing the data with sufficient accuracy. Many machine learning practitioners
believe that properly optimized feature extraction is the key to effective model construction. Determining a
subset of the initial features is called feature selection. The selected features are expected to contain the
relevant information from the input data, so that the desired task can be performed by using this reduced
representation instead of the complete initial data. In this module, we can select the many attributes from pre-
processed datasets.
ALGORITHM
Artificial Neural Network Algorithm algorithm is a supervised classification algorithm. We can see it from its
name, which is to create a forest by some way and make it random. There is a direct relationship between the
number of trees in the forest and the results it can get: the larger the number of trees, the more accurate the
result. But one thing to note is that creating the forest is not the same as constructing the decision with
information gain or gain index approach. The decision tree is a decision support tool. It uses a tree-like graph to
show the possible consequences. If you input a training dataset with targets and features into the decision tree,
it will formulate some set of rules. These rules can be used to perform predictions.
Classification
Classification in this system refers to the process of categorizing users based on their input parameters to
generate appropriate diet recommendations using an Artificial Neural Network (ANN). The System operates
within a Machine Learning Environment, processing user data to generate recommended diet plans. The
dataset is divided into three categories:
1. Lunch Data
2. Breakfast Data
3. Dinner Data
DIET RECOMEMDATION
• Recommended caloric and nutritional needs
• Suggested diet plan
V. RESULTS AND DISCUSSION
The implementation of the Personalized Diet Recommendation System using Artificial Neural Networks
demonstrated promising results, with the model accurately generating diet plans tailored to individual user
profiles based on inputs such as age, BMI, activity level, and dietary preferences. During testing, the ANN
e-ISSN: 2582-5208
International Research Journal of Modernization in Engineering Technology and Science
( Peer-Reviewed, Open Access, Fully Refereed International Journal )
Volume:07/Issue:07/July-2025 Impact Factor- 8.187 www.irjmets.com
www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science
[2144]
achieved high prediction accuracy in aligning recommendations with user health goals, such as weight loss or
muscle gain, and showed strong adaptability when user inputs were updated. User feedback indicated
increased satisfaction and adherence to the suggested plans, highlighting the effectiveness of the system’s
personalization. The results confirm that ANNs can efficiently process complex, multidimensional health data to
provide dynamic and relevant dietary guidance, marking a significant improvement over traditional static diet
plans.
Figure 2:
NUTRIENTS INTAKE SUGGESTION
Figure 3:
In this page, suggested nutrients value will be displayed
VI. CONCLUSION
In this project, a Personalized Diet Recommendation System was developed using Artificial Neural Networks
(ANNs) to address the growing demand for individual-specific nutritional guidance. By collecting essential
health and lifestyle inputs such as age, gender, BMI, dietary preferences, exercise intensity, weight goals, and
food allergies, the system intelligently predicts the user's nutritional needs and offers customized food
e-ISSN: 2582-5208
International Research Journal of Modernization in Engineering Technology and Science
( Peer-Reviewed, Open Access, Fully Refereed International Journal )
Volume:07/Issue:07/July-2025 Impact Factor- 8.187 www.irjmets.com
www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science
[2145]
recommendations. The project successfully demonstrates that machine learning models, particularly ANNs, can
effectively analyze user data and provide accurate dietary suggestions that align with personal health goals.
VII. REFERENCE
[1] S. D. Luhach, K. Kharb, A. Saxena, and K. Saini, “Food recommendation system for diabetic patients
using machine learning,” IEEE International Conference on Computing, Communication and
Automation (ICCCA), 2020.
[2] P. Singh, N. Joshi, and R. Patil, “AI-Based Personalized Diet Planner Using Nutritional Ontology and
ANN,” International Journal of Advanced Computer Science and Applications (IJACSA), vol. 11, no. 6, pp.
420–427, 2020.
[3] M. Sharma, S. Kumar, and V. Sharma, “Nutritional food recommender system using hybrid filtering
techniques,” International Conference on Intelligent Computing and Control Systems (ICICCS), 2021.
[4] X. Liu, X. Yuan, Y. Zhao, and L. Wu, “A personalized diet recommendation system based on health
monitoring data,” IEEE Access, vol. 8, pp. 10385–10396, 2020.
[5] A. Pandey, N. Srivastava, and T. Kapoor, “Personalized meal recommendation for healthy living using
deep learning,” International Journal of Computer Applications, vol. 182, no. 23, pp. 25–30, 2020.
[6] T. Nguyen and D. Nguyen, “Smart nutrition recommendation system using machine learning
algorithms,” Proceedings of the 5th International Conference on Green Technology and Sustainable
Development, 2021.
[7] M. S. Hossain, M. A. Rahman, and M. S. Kaiser, “An intelligent healthcare recommendation system for
personalized diet planning using deep learning,” IEEE Reviews in Biomedical Engineering, vol. 14, pp.
175–188, 2021.
[8] J. A. Thomas and M. S. Anand, “Food recommendation system using deep neural networks,” Journal of
Artificial Intelligence and Soft Computing Research, vol. 10, no. 2, pp. 55–62, 2020.
[9] S. Bhatia and A. Sinha, “Dietary recommendation engine using machine learning for lifestyle disease
management,” International Journal of Scientific & Technology Research, vol. 9, no. 1, pp. 313–318,
2020.
[10] G. J. Kloss, “Personalized nutrition and artificial intelligence: An approach to preventive health,”
Nutrients, vol. 13, no. 2, pp. 589–599, 2021.
