Original scientific paper
A Financial Data Mining Model for Extracting Customer Behavior
Mark K.Y. Mak
; Convoy Financial Services Holdings Limited, Hong Kong
George T.S. Ho
; Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University, China
S.L. Ting
; Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University, China
Abstract
Facing the problem of variation and chaotic
behavior of customers, the lack of sufficient information
is a challenge to many business organizations. Human
analysts lacking an understanding of the hidden patterns
in business data, thus, can miss corporate business
opportunities. In order to embrace all business
opportunities, enhance the competitiveness, discovery of
hidden knowledge, unexpected patterns and useful rules
from large databases have provided a feasible solution for
several decades. While there is a wide range of financial
analysis products existing in the financial market, how to
customize the investment portfolio for the customer is
still a challenge to many financial institutions. This paper
aims at developing an intelligent Financial Data Mining
Model (FDMM) for extracting customer behavior in the
financial industry, so as to increase the availability of
decision support data and hence increase customer
satisfaction. The proposed financial model first clusters
the customers into several sectors, and then finds the
correlation among these sectors. It is noted that better
customer segmentation can increase the ability to identify
targeted customers, therefore extracting useful rules for
specific clusters can provide an insight into customers’
buying behavior and marketing implications. To validate
the feasibility of the proposed model, a simple dataset is
collected from a financial company in Hong Kong. The
simulation experiments show that the proposed method
not only can improve the workflow of a financial
company, but also deepen understanding of investment
behavior. Thus, a corporation is able to customize the
most suitable products and services for customers on the
basis of the rules extracted.
Keywords
Association Rules Mining; Clustering; Customer Behavior; Data Mining; Financial Industry
Hrčak ID:
71515
URI
Publication date:
15.8.2011.
Visits: 4.406 *