Document Type
Honors Thesis
Abstract
This study explores the effectiveness of a hybrid recommendation system for e-commerce by integrating content-based, collaborative, and popularity-based models. Traditional individual algorithms have inherent limitations, such as handling new users or items data sparsity and ensuring relevance and diversity in suggestions. The hybrid model seeks to overcome these challenges by leveraging the strengths of all three methods, thus potentially offering more precise, personalized product suggestions. The performance of each model and its integration into a hybrid system are evaluated through logistic regression analysis. Initial results indicate that the hybrid system significantly outperforms the individual models in terms of accuracy and user satisfaction. This research underscores the potential of hybrid recommendation systems to enhance user experience and support businesses in optimizing their online platforms. Importantly, the study provides practical insights for e-commerce enterprises aiming to refine their customer interaction strategies, demonstrating the real-world relevance of the research.
Disciplines
Computer Engineering
Publication Date
5-2024
Language
English
Faculty Mentor of Honors Project
Christopher Conly
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
Recommended Citation
Safat, Saniah, "Effectivity Analysis of Recommendation Algorithm: A Comparative Study of the Performance of a Hybrid Model and an Individual Recommendation Algorithm" (2024). 2024 Spring Honors Capstone Projects. 4.
https://mavmatrix.uta.edu/honors_spring2024/4