Apparel Recommendation System using Rule based Filtering and SMOTE for Online Shopping

The Apparel Recommendation System (ARS) is a machine learning-based platform designed to enhance the shopping experience by providing personalized apparel suggestions based on user preferences such as type, color, price range, and ratings. The user interface, developed with Tkinter, is simple and in...

Full description

Saved in:
Bibliographic Details
Published in2025 International Conference on Artificial Intelligence and Data Engineering (AIDE) pp. 595 - 601
Main Authors Kamath, B Sunil, Alva, Rashmith R, Shetty, Ridhi, Dinakar, Sharanya, Shetty, Sakshi
Format Conference Proceeding
LanguageEnglish
Published IEEE 06.02.2025
Subjects
Accuracy
Apparel recommendation
Fatigue
Filtering
Image color analysis
Impedance matching
Machine learning
Real-time systems
Recommender systems
Rule-based filtering
Shopping
SMOTE
User experience
User interfaces
Online AccessGet full text
DOI10.1109/AIDE64228.2025.10987346

Cover

More Information
Summary:The Apparel Recommendation System (ARS) is a machine learning-based platform designed to enhance the shopping experience by providing personalized apparel suggestions based on user preferences such as type, color, price range, and ratings. The user interface, developed with Tkinter, is simple and interactive, allowing effortless selections through drop-down menus and sliders. On the backend, SMOTE (Synthetic Minority Over-sampling Technique) addresses dataset imbalances, ensuring fair representation across apparel categories. A rule-based filtering approach aligns recommendations with user-selected preferences, resulting in accurate suggestions. Additionally, a live chat feature, integrated with React.js, Node.js, and Socket.io, offers real-time assistance. The curated dataset improves recommendation accuracy, enabling users to quickly find apparel that matches their individual styles.
DOI:10.1109/AIDE64228.2025.10987346