www.rsisinternational.org
Page 4019
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)
ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue IX September 2025
Estilo: A Mobile-Based Fashion Recommendation App Tailored to

Boquiron, Carl Leonard*, Cabrito, John Adrian, Battad, Zydane Diesel, Brisenio, Mathew, Fernandez,
Ronald
1* College of Computing Studies, Universidad De Manila, Philippines
*Corresponding Author:



ABSTRACT
Artificial Intelligence (AI) and Machine Learning (ML) have been increasingly affecting the fashion industry
with real-time capabilities for personalized clothing. Despite fashion ultimately being just self-expression,
identity & culture, many considerations, like differences in body shapes, tastes, requirements for the occasion,
and continuously evolving trends, complicate the decision-making on what to wear. While personalization
systems are more beneficial than one-size-fits-all, many systems still limit their capability for personalization
through apps that do not focus on user preference and that are less accurate. This study shows the rapid evolution
of a cross-platform mobile application for fashion recommendations, that fused content-based and Collaborative
filtering with machine learning techniques to deliver personalized fashion recommendations. The app simplifies
the choices about outfitting, reduces the time browsing for clothing, builds user confidence, and provides several
options suitable to various user preferences. Firebase and Supabase offer database management and
authentication security, while Machine Learning is leveraged to analyze the strong relationships between user
and product data in order to provide a recommendation based on user preferences. The development utilized
Agile methodologies, incorporating iterative tasks and adjustments guided by user feedback to improve
functionality, usability, and precision. Result demonstrates that the application reduces time and browsing in
finding outfits and increase user confidence through reliable, and timely suggestions suited to the situation.
Moreover, the system showcased inclusivity by putting various styles and real-time trends, thus enabling
merchants to connect with a wider audience. In summary, the findings shows that an AI-mobile based fashion
recommendation system provides a more user-friendly, personalized, and various clothing selection method. The
suggested solution promotes digital fashion technologies by focusing on individuality, diversity, and usability,
thereby improving the role of AI in daily self-expression.
Keywords: Fashion, Recommendation System, Content-based Filtering, User Preference, Personalized Fashion
INTRODUCTION
Information technology influences the personal and professional lives of each individual in modern society.
Information technology forms the baseline for all technology advances and inventions, affecting the present and
future. It is known as artificial intelligence (AI) when an information technology application can perform
repetitive activities requiring human intelligence. A division of AI, machine learning (ML) is typically defined
as an application of algorithms to develop, improve, and predict applications such as recommendation systems.
With the help of Artificial Intelligence (AI) and Machine Learning (ML), the system provides users with
personalized outfit by removing the uncertainty associated with what to wear to a specific preference of a user.
Instead of relying solely on user preferences, it makes recommendations based on the characteristics of the items.
The recommendations are coordinated based on the attributes to meet the needs of the end user. It can also
increase convenience and save time by giving the end-user precise clothing options. Additionally, the system
adapts to the current trends and allow the end-user to stay updated with the evolution of fashion, offering
coordinated style that enhance confidence in their appearance.
Choosing the appropriate outfit and making sure it suits the style and personality of a person are some of the
most difficult fashion challenges. According to Dabare and Caldera (2024), by examining the end-user
www.rsisinternational.org
Page 4020
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)
ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue IX September 2025
preferences and wardrobe items, AI-driven recommendation systems can assist end-users in making better
fashion choices. By utilizing machine learning algorithms to provide personalized outfit suggestions.
Therefore, this study seeks to develop Estilo: A Mobile-based Fashion Recommendation App Tailored To User's
Need. The goal of this system is to make a list of suggestions for the end-user without making them take the
guesswork and anxiety out of choosing what to wear. It is designed for making it easier to look for outfits that
is most suitable based on the choices of the user. Feeling confident while being fashionable with minimal effort,
enhancing their daily- decision without worrying about stress.
REVIEW RELATED STUDIES
Sivaranjani, L., et al. (2023) express that AI-based applications can help to personalize outfit suggestions based
on user preferences. This reinforces that both machine learning and content-based filtering can be beneficial
technologies in recommender systems based on the prior behavior of the user. In turn, this increases the
personalization aspects and efficiency in the selection of wearables in mobile applications.
A study by Varma et al. (2024) they developed a system STYLESYNC, a fashion recommendation system that
uses AI-Based to assist an individual to decide what to wear. The interface of the STYLESYNC is user friendly
allowing the individual to conveniently coordinate their clothing items in their wardrobe and affords the use of
the feedback supplied by users to continue improving the accuracy of the recommendations and user experience.
This study reflects the capabilities of AI for fashion recommendations.
Khubchandani et al. (2024) develop a Smart Outfit Recommendation System that provides clothing options
based on the user's body type, gender, skin tone, and special occasions. With an intuitive interface, the system
provides similies about the user's wardrobe for outfit assembling and includes real-time interactive fashion
recommendations. This study highlights the importance of fashion recommendations based on user behavior and
has expanded this idea further by providing handy content-based filtering in outfit suggestions that shows the
users' own wardrobe.
According to Blancaflor and Villasor (2025), they developed a fashion e-commerce with a recommendation
system and augmented with virtual reality (VR) for a more enjoyable buying experience. The utilization of AI
recommendation system and emerging technologies leads to personalization in shopping experiences. The results
of this study contribute to the development of a mobile system for recommending outfits by showing that
recommendation algorithms and virtual try-on increase user engagement and facilitate better outfits selections
suited to individual preferences.
Mire (2021), developed an e-commerce application with a recommender system. Its job is to provide users with
personalized product recommendations based on their ratings, which indicate their likes and dislikes or what
they preferred. It emphasizes a recommendation system for user action through AI-driven personalized
recommendations.
According to Yaswanthraj et al. (2024), they developed a comprehensive clothing recommendation system that
can highlights the important role in enhancing deep levels of user satisfaction and efficiency within the online
shopping experience. The study study focused on creating that the addition of user attributes to outfit
recommendations, example of that is body shape, skin tone, height, and style preference, allows the system to
deliver more precise and personalized recommendations through user characteristics and preferences.
According to Sarojadevi et al. (2023), product recommendation engines are effective tools in improving the
quality of customer experience through the use of machine learning, various product selection forms, and a
specialized recommendation system algorithm. The researchers in the study titled Fashion Recommender System
(FRS): Image-Based Engine for Personalized Outfit note that fashion recommendation systems dynamically
integrate into various platforms, including websites, applications, and digital platforms. They stated that current
engines can also process large product catalogues and autonomously determine which algorithm(s) and filters to
use for each individual user. Due to this, businesses can enhance personalization, improve conversion rates, and
increase customer satisfaction overall.
www.rsisinternational.org
Page 4021
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)
ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue IX September 2025
According to Challa et al. (2023) An intelligent system called a fashion recommendation system provides
customers personalized fashion recommendations based on different factors including their own preferences,
style, body type, and other criteria. The system processes the user’s data and recommends the most suitable
fashion products for them using data analytics, machine learning, and artificial intelligence approaches.
Intelligent fashion recommendation is urgently needed due to the tremendous growth of fashion-oriented trends.
We developed algorithms that automatically recommend the users’ clothing choices based on their fashion
preferences.
Becattini et al. (2023) suggest that improving fashion recommendations looks at a new interpretation of "layer,"
which expresses and incorporates not only style but also social context. In the study, Fashion Recommendation
Based on Style and Social Events, they represented style as a hybrid of mood and emotional attributes of color
combinations, while using social context as a measure of suitability for an outfit via analysis of actual social
context or events of clothing worn in images. By incorporating this representation of style and context into their
recommendation framework, they first implemented a style classifier and then a context classifier to make
contextual fashion recommendations.
According to Guillermo et al. (2021), recommender systems are becoming increasingly important as most
software applications are aggregated, giving a one-stop shop for clients, particularly in the fashion industry.
Based on the data and findings of the study presented in the preceding section, it is possible to conclude that the
demonstration of unsupervised picture learning was successful. Aside from practicality, aggregated systems will
be more efficient and user pleasant in the sense that a user will not have to walk through the complete categories
of an online shop or even search from multiple applications in order to find the item that is being searched for.
CONCEPTUAL FRAMEWORK
The conceptual framework of the study “Estilo, A Mobile-Based Fashion Recommendation App Tailored to
User's Need” is made including the input, output, and processes, to obtain the overall picture of the study. The
goal of this study is to find what the end-user wants for their apparel on a certain occasion. The following figure
is shown as the conceptual framework:
Figure 2.1: input-process-output model (IPO)
www.rsisinternational.org
Page 4022
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)
ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue IX September 2025
The figure 2.1 shows the stages that the study needs to take in order for the proponents achieve its desired output,
and that will be the mobile fashion recommender with a smart filtering algorithm in order to sort what the user
likes.
Input stage are consists of hardware, software requirements, user preferences, and apparel specifications.
Hardware requirements include the laptop, computer, and a mobile phone used for designing, developing, testing
of the mobile application. For the software requirements, it includes React Native (JSX) as the framework for
mobile development and Python for handling the database, with an IDE such as Visual Studio Code or Android
Studio used for coding and development. For the user requirements, it includes the what the user like when
picking an outfit. Lastly for the apparel specifications, this includes the size, type of attire, and the type of event
that the end-user will filter out in the application.
Process stage it includes the planning, designing, developing, testing, deploying, and the review. In the first
phase of the Agile methodology will be the data collection, it is the phase wherein the different data or
information will be gathered for the study. Afterwards, the data analysis will be the next phase wherein this will
be the phase that we filter out the necessary data that is needed for the study. Following that, will be the designing
phase which will be the UI/UX of the mobile application. Next, will be the developing phase, this is the phase
that will focus on coding, where the source code will be made so that the application will be functional. Next
will be the testing phase, this phase will focus on debugging and testing in order for the application to have
minimal errors. Next is the deployment phase where the application will be available for use on Android. Lastly,
the review, where feedback from users will be gathered to evaluate the performance of the application and in
order for the proponents to identify the areas for improvement.
Output stage will include what will be the end product or the mobile application itself, while including the user
feedback that comes from the application. The mobile application “Estilo” that has been built, will be ready for
deployment and functional with all its core features, ensuring a seamless user experience. In the user feedback,
it will be used for improvements for the application in order for it to be more efficient and is aligned to meet the
preference of the user.
METHODOLOGY
The process of developing Estilo" A Mobile-Based Fashion Recommendation App Tailored to User's Need. It
uses the Agile Model Development that makes sure that the mobile application develops continuously for every
stage and requires user feedback for further improvements. The Agile methodology was broken down into
iterative cycles of planning, design, development, testing, release, and feedback. Each sprint ended with user
evaluations to identify issues and measure usability and accuracy for ongoing improvement. Agile methodology
is flexible that helps Estilo grow and evolve. The experimentation we can conduct, the data we collect, and the
recommendations we are able to refine create a more intelligent and engaging experience for the user. Whether
it is a small change in the look and feel or the larger changes to the machine learning models designed to deliver
on the outfit recommendations, being agile allows for the building of a product that meets the expectations of
the end-users while reducing the exponential rate of change within the fashion space.
Figure 3.1: Agile Model
www.rsisinternational.org
Page 4023
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)
ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue IX September 2025
Planning
This stage will involve gathering the essential data that will be needed for the fashion recommendation system.
The proponents will conduct surveys, interviews, and gather data sets. The proponents will be using these data
gathering techniques to identify all user preferences, challenges, and inspirations for our project. The necessary
data will be analyzed for the proponents to determine the key factors influencing apparel recommendations, such
as fashion trends, personal style preferences, clothing suitability for different weather conditions.
Design
The proponents will design the user interface (UI) and user experience (UX) using Figma. Wireframes and
prototypes will be created for users for the proponents to visualize how users interact with the application. In
order to align the content for
In order to fit the design with the user's preferences. The proponents conduct an interview with a specific
individual to gain information into how they felt after using the app, the content, the design, the color, and the
outfit recommendations. Proponents ensure that it aligns with the user's preferences.
Develop
The proponents will use react native as Integrated Development Environment (IDE) for the availability in
Android, using JavaScript to build the Frontend and python for the Backend, will also be using Firebase and
Supabase for the database management.
To improve the recommendation system, the algorithms were tested on a sample collection of outfit photographs
organized by size, event, weather, and style. The models were created in Python and linked to frontend
processing via React Native using Firebase and Supabase databases. Precision and recall metrics were used to
assess the correctness and relevance of recommendations.
The proponents begin by identifying the necessary datasets, in this case, the outfit for the user needs to be find.
www.rsisinternational.org
Page 4024
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)
ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue IX September 2025
Furthermore, the proponents create the basis and fundamental frameworks, such as Content-Based Algorithm
and Collaborative Filtering to work them together to be able understand the user preference, because these are
required for the recommender system to function. Using the essential frameworks, the system will be able to
understand user actions, making outfit suggestions more relevant and assisting users in discovering new outfit
ideas.
Test
Testing will be carried out to locate and fix bugs. Unit, integration, and user acceptance testing (UAT) is
conducted so that it will assess system performance, accuracy, and usability. The testers will check the accuracy
of recommendations and speed across various devices and screen sizes.
The proponents have gathered local apparel businesses that will serve as the study's respondents, including Street
Underground in Paco, Tondo Manila, Yifu in Binondo Manila, and Plain Street in Tondo Manila. The proponents
create a Likert-Scale interview questionnaire for the intended target user’s to review and test the application.
The questionnaire was created using the ISO/IEC 25010 criterion to ensure the applications fit the user's wants
and needs and also to assess the quality standard, functionality, efficiency, and reliability.
Release
The application will be deployed in a live setting. The deployment includes provisioning cloud infrastructure,
considerations of data security and operational availability through app distribution services. The team will
provide feedback channels for early adopters and stakeholders to help with refinement based on feedback.
Feedback
The proponents will use the ongoing review and feedback process to further develop the system. The proponents
will release regular updates to provide improvements to prompt features, updates of recommendation accuracy,
and apply any lessons learned. Review of consumer updates are analyzed for priority of features in future
releases, and to ensure the app remains with fashion trends and preferences of the users.
The proponents will gather and document performance evaluation data, such as accuracy, precision rate, and
response time, after each development sprint to ensure transparency and reproducibility. User feedback guided
the changes to the UI and algorithm of the system, allowing for more flexible and user-centered fashion
recommendations.
With this, the proponents will assess system performance based on user feedback and make any necessary
changes and upgrades to preserve the quality of outfit recommendations.
SAMPLING METHODS
The study is conducted with fashion enthusiasts, fashion designers, and fashion retailers, they considered to be
the primary users of the system. The purposive sampling technique is designed to select individuals with relevant
expertise and knowledge in the fashion industry.
The purposive sample was chosen to guarantee that the participants in the study had the ability to provide useful
www.rsisinternational.org
Page 4025
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)
ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue IX September 2025
data and information. The goal was to collect in-depth information from users in order to support and contribute
to the early stages of study development and analysis.
ETHICAL CONSIDERATION
The proponents followed to ensure the integrity, safety, and fairness of the study. It addresses the protection and
privacy of the information that obtained to the study. In conducting this study, proponents ensure the integrity
and fairness of the AI-based outfit recommendation system without bias, making sure that AI-based outfit
recommender systems are fair and inclusive to anyone and protect the privacy of users and make sure it is
encrypted while ensuring that the data collected and recommendation that the user personalized are conducted
transparently and securely. Additionally, sensitive information was protected against unauthorized access, and
academic integrity was upheld by properly citing all sources used in the study.
RESULT AND DISCUSSION
The survey results obtained from the total 60 respondents regarding the evaluation of the mobile-based fashion
recommendation application, Estilo. The data collected pertained to six quality criteria, namely functional
suitability, performance efficiency, usability, reliability, security, and portability. The results are presented in
tables accompanied by mean scores and were interpreted based on a Likert scale (1=Strongly Disagree to
5=Strongly Agree) and a graphical scale for a better result perception. The subsequent discussion draws attention
to how these results align with the aims of the research study.
RESULT
Functionality
Figure 4.1: Pie Graph for Functionality
www.rsisinternational.org
Page 4026
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)
ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue IX September 2025
The system received an average mean score of 4.24 (Very Good) regarding functional suitability. The majority
of respondents agreed the application has the appropriate functions and is providing adequate and accurate outfit
suggestions according to stated preferences.
This shows how well the functions of the system respond to the user's needs. Additionally, it provides further
evidence that ESTILO, as a mobile-based fashion recommendation application, provides sufficient and accurate
clothing recommendations that are compatible with the preferences of the user's intended audience.
PERFORMANCE
Figure 4.2: Pie Graph for Performance
The system achieved an average mean score of 4.03 (Good) for performance efficiency. Users generally agreed
that the application loads quickly, responds promptly, and handles image-heavy content efficiently with smooth
interactions.
The measurement of performance efficiency shows that the system was able to provide sufficiently fluid and
resonant interactions. Additionally, the results also demonstrates that ESTILO, as a mobile-based fashion
recommendation application, has great loading times, as well as being able to successfully render content that is
image heavy, which allows for an affectively positive user experience.
Usability
Figure 4.3: Pie Graph for Usability
www.rsisinternational.org
Page 4027
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)
ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue IX September 2025
The system received a mean score of 4.33 (Very Good) for usability. Most of the respondents strongly agreed
that the application is easy to navigate utilizing clear icons and instructions, requiring very little effort to learn.
The usability results indicate that the system is straightforward and intuitive for users to control. Furthermore, it
is further evidence that ESTILO provides clear icons as well as clear instructions about the process, making
navigation simple and requiring very little cognitive effort, as shown by its mean score of 4.33 (Very Good).
Reliability
Figure 4.4: Bar Graph for Reliability
The system was able to gain a mean average score of 3.98 (Good) in survey. The users concurred that the
application rarely crash, works in different conditions of internet and maintains data persistence effectively.
The results show that the system works reliably under various usage conditions. ESTILO operates reliably and
stably, rarely crashing, and can operate with changes in the internet service. It also allows users to be dependent
on it to use it in consistently and safer way because of data persistence.
Security
Figure 4.5: Bar Graph for Security
www.rsisinternational.org
Page 4028
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)
ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue IX September 2025
The system achieved an average mean score of 4.13 (Good) regarding security. Respondents agreed that the
application adequately protects personal data, provides privacy control, and ensures account safety.
The investigation reveals that the system provided adequate protections for users’ information. ESTILO accounts
are safe, personal data is secured and privacy controls allow the user to easily be aware of how safe the
application and their security is.
Portability
The system received an average mean score of 4.23 (Very Good) for portability. Users agreed that the application
offers easy installation, works across different devices, and adapts well to varying screen sizes.
Figure 4.6: Bar Graph for Portability
The results indicate that the system is adaptable and easy for users to access across platforms. The ESTILO
system installed easily, worked on devices and adjusted nicely for different screen sizes, and made access to
system and usage easy and user-friendly in a variety of using conditions.
User Evaluation Survey Summary
Overall, the system received positive reviews across all six rating scales, with usability and functional suitability
receiving the highest scores. These findings indicate that the application achieves user expectations by offering
accurate and individualized recommendation sets (outfits) while also providing a simple, appealing, and user-
friendly interface. The neutral to favorable evaluations for performance and reliability indicate that there are still
areas for us to improve optimization, particularly in terms of reliability.
www.rsisinternational.org
Page 4029
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)
ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue IX September 2025
Table 1: User Evaluation Survey Summary
CONCLUSION
In summary, this study is designed to developed a successful mobile application that integrated with AI-powered
outfit recommendation system. The structured development of the system will go to the process of agile
methodology that breaks the project down into several and manageable phases that would help to the
development of the project and can also adapt for a continuous improvement by utilizing artificial intelligence
and machine learning models, the proponents can achieve the project goals of this study to provide system that
can personalized outfit suggestion based on user preferences by integrating Content-based filtering, and Random
Forest with this the system can analyze various outfit attributes to recommend a outfit based on the user selected
and to enhance the accuracy and the performance of the system to recommend. Through the process, the
proponents gathered data, such as surveys and questionnaire to a certain individual that can help and give their
insight to the development of the project and also the proponents collect a literatures data that have connection
to the study for a support. The development process of the project will cover the model training, gathering data,
system design, testing and evaluation to ensure the accuracy of the recommendation and the responsiveness of
it in mobile application that align to the project goal of the proponents to develop a mobile-based outfit
recommendation for personalized choices.
RECOMMENDATIONS
Base on the result of the study, the researcher recommends the following for the better improvement in the future
of Estilo: A Mobile-based Fashion Recommendation App Tailored to User’s Need”
www.rsisinternational.org
Page 4030
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)
ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue IX September 2025
Integration of Camera for Outfit Suggestion- clothes suggestion through built-in camera using CNN
algorithm.
Incorporation of lightning in Outfit Suggestion Recommends outfit choices suited to specific lighting
situations. Users are to capture or photograph the lighting conditions of the wearer's environment.
Messaging System - This feature allow users to communicate without the need of external application,
enhancing user convenience, streamline interaction, and engagement within the application.
Real-time Weather Integration – Incorporating real-time weather data would enhance the accuracy of outfit
recommendation.
REFERENCES
1. Dabare, A. N. H., & Caldera, H. I. A. (2024, November). Wear Me: Clothing Feedback and
Recommendation Mobile App. In 7TH INTERNATIONAL CONFERENCE ON BUSINESS
INNOVATION (1), (p.302).
https://www.researchgate.net/publication/387318311
2. Sivaranjani, L., Rachamadugu, S. K., Reddy, B. S., Sakthivel, M., & Depuru, S. (2023, September).
Fashion recommendation system using machine learning. In 2023 4th International Conference on Smart
Electronics and Communication (ICOSEC)(2), (pp. 1367-1374). IEEE.
DOI:
10.1109/ICOSEC58147.2023.10275967
3. Varma, N., Singh, G., Singh, V. R., Swami, V., Tyagi, A., & Verma, S. (2024). STYLESYNC: A fashion
recommendation system. International Journal of Scientific Research in Engineering and Management,
8(3), (5), 1–5. https://doi.org/10.55041/ijsrem34672
4. Khubchandani, P., Gavali, V., Ghanwat, H., Gawali, N., Gautam, M., & Gawade, S. (2024, December). A
Web-Based Smart Outfit Recommendation System Using Node. js, Express. js and MySQL for Dynamic
Fashion Recommendations. In 2024 International Conference on Sustainable Communication Networks
and Application (ICSCNA) (4), (pp. 576-579). IEEE. DOI:
10.1109/ICSCNA63714.2024.10864223
5. Blancaflor, E., & Villasor, D. A. (2024). StyleSavvy: A Design of a Personalized AI-Driven Personal
Shopper with a Virtual Try-On Feature. Proceedings of the 2024 7th International Conference on
Information Science and Systems, (5),129–133.
https://doi.org/10.1145/3700706.3700728
6. Birwadkar, S., Firke, R., Raje, V., Vaity, R., Mire, A., & Student, B. (2021). Fashionista: A Clothing
Application with Collaborative Filtering Based Recommendation System. 8. (6)
https://d1wqtxts1xzle7.cloudfront.net/87863209/JETIRFD06037-libre.pdf?1655873216
7. Yaswanthraj, Nadar, P. K., Moorthy, M. S., & Sivakumar, S. (2024, December 19). AI-BASED OUTFIT
RECOMMENDATION SYSTEM. ResearchGate.(7) https://doi.org/10.13140/RG.2.2.26202.25287
8. Sarojadevi, H., Bhat, V., Palekar, R., & Neha, K. (2023, April) Fashion Recommender System (FRS):
Image Based Engine for Personalized Outfit. Vol.12 No.1, 2023, pp.21-24 (8)
DOI: https://doi.org/10.51983/ajcst-2023.12.1.3569
9. Challa, N. P., Abbaraju Sao Sathwik, Kiran, J. C., Kokkula Lokesh, Ch, D., & Beebi Naseeba. (2023).
Smart Fashion Recommendation System using FashionNet. ICST Transactions on Scalable Information
Systems.(9) https://doi.org/10.4108/eetsis.4278
10. Becattini, F., Divitiis, L. D., Baecchi, C., & Bimbo, A. D. (2023). Fashion recommendation based on style
and social events. Multimedia Tools and Applications, 82(24), 38217–38232.(10)
https://doi.org/10.1007/s11042-023-15290-4
11. Guillermo, M., Espanola, J., Billones, R. K., Vicerra, R. R., Bandala, A., Sybingco, E., Dadios, E. P., &
Fillone, A. (2021). Content-based Fashion Recommender System Using Unsupervised
Learning. TENCON 2021 - 2021 IEEE Region 10 Conference (TENCON), 29–34.(11)
https://doi.org/10.1109/tencon54134.2021.9707459