Year: 2025
Type: article
Abstract: A financial product recommendation system is an important tool for financial institutions in financial services to enhance customer experience and satisfaction. Such systems provide personalized recommendations, assisting users in identifying and selecting appropriate financial products. However, traditional recommendation methods, including collaborative filtering and content-based recommendation, frequently suffer from limited personalization capabilities due to challenges like sparse data and user cold-start problems. This study investigates the application of cluster analysis in financial product recommendation systems and the role of various clustering algorithms (e.g., K-means, DBSCAN, and hierarchical clustering) in customer segmentation and personalized recommendations and evaluates their advantages and challenges in optimizing recommendation strategies. The results show that clustering analysis can effectively group users and products, significantly improving product matching and, thus, recommendation accuracy and customer satisfaction. In addition, this paper discusses the main challenges of applying clustering techniques to financial recommendation systems, such as ensuring the scalability of clustering models and accurately evaluating validation clustering results.
Author: Ni Rui
Institution: Hong Kong Baptist University
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4410129326