Deep Learning-Driven Personalized Recommender Systems: Theory, Models, and Future Directions

Xingzhe Feng, Ziyi Sheng

2024

Abstract

Deep learning-based recommendation algorithms have emerged as a significant area of interest within artificial intelligence research. This surge in attention is primarily due to the limitations of conventional recommendation systems, coupled with the rapid advancements in deep learning technologies. In this work, we provide an overview of deep learning-based recommender systems, including their key stages and components. The pipeline will be explicated, specially, data collection and preprocessing, feature engineering, model selection and training, evaluation and the deployment with online learning mechanisms would be stressed. Additionally, we introduce a novel deep learning-based recommender system named Stratified Advance Personalized Recommendation System (SAP Model). This system solves the problem in the recommendation of the cold start, overspecialization, and data sparsity. By Stratified, we mean that this method personalizes the recommendation by clustering the users who have similar interactions or demographics. The architecture, training techniques, and evaluation measures of the SAP Model will also be covered, which gives us a glance at the improvement of the effectiveness of recommendations and the satisfaction of users in the real world.

Paper Citation

in Harvard Style

Feng X. and Sheng Z. (2024). Deep Learning-Driven Personalized Recommender Systems: Theory, Models, and Future Directions. In Proceedings of the 1st International Conference on Engineering Management, Information Technology and Intelligence - Volume 1: EMITI; ISBN 978-989-758-713-9, SciTePress, pages 105-109. DOI: 10.5220/0012910700004508

in Bibtex Style

@conference{emiti24,
author={Xingzhe Feng and Ziyi Sheng},
title={Deep Learning-Driven Personalized Recommender Systems: Theory, Models, and Future Directions},
booktitle={Proceedings of the 1st International Conference on Engineering Management, Information Technology and Intelligence - Volume 1: EMITI},
year={2024},
pages={105-109},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012910700004508},
isbn={978-989-758-713-9},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Engineering Management, Information Technology and Intelligence - Volume 1: EMITI
TI - Deep Learning-Driven Personalized Recommender Systems: Theory, Models, and Future Directions
SN - 978-989-758-713-9
AU - Feng X.
AU - Sheng Z.
PY - 2024
SP - 105
EP - 109
DO - 10.5220/0012910700004508
PB - SciTePress