Enhanced Multimodal Recommendation System for Personalized Lifestyle Recommendations

DOI

10.2991/978-94-6463-700-7_5How to use a DOI?

Keywords

Hybrid Recommendation; Personalized; Comparative Study

Abstract

The system learns to object similarities by contrasting loss based on multimodalities with data of similar and dissimilar objects under consideration for a recommendation. This research provides a multimodal recommendation with an advanced framework for personalized lifestyle recommendations that fabricates a model that incorporates numerous modalities such as images, text, and user interaction data. It includes components such as preprocessing, neural image feature extraction, and embedding of the text using Natural Language Processing models like BERT. The fulfillment and inclusion of an attention mechanism fuses these features into a more complete multimodal representation that yields better performance in terms of recommendation accuracy. Further optimization strategies like Adam and regularization techniques (L2, L1) are used to maintain stability and avoid overfitting of the model. This study presents a hybrid recommendation approach that has incorporated Content-Based Filtering with the Collaborative Filtering method for a relevant recommendation process. Contrastive learning ensures that it captures subtle differences between similar items, resulting in a more varied and accurate performance of recommendations. System performance is evaluated in several scenario settings to depict an improvement in user-friendliness, recommendation accuracy, and diversity over traditional techniques. It summarizes such a way with a cognition concerning personalized digital experiences into lifestyle recommendation by pushing forward the application around multimodal fusion and contrastive learning. Such applications transcend to fashion, health, entertainment, and other potential areas.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Mahima Kansal
AU  - Sohit Agarwal
PY  - 2025
DA  - 2025/04/19
TI  - Enhanced Multimodal Recommendation System for Personalized Lifestyle Recommendations
BT  - Proceedings of the International Conference on Advancements in Computing Technologies and Artificial Intelligence (COMPUTATIA-2025)
PB  - Atlantis Press
SP  - 36
EP  - 51
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-700-7_5
DO  - 10.2991/978-94-6463-700-7_5
ID  - Kansal2025
ER  -