An integrated framework for recommendation systems in e‐commerce
| Date | 01 November 2002 |
| DOI | https://doi.org/10.1108/02635570210445853 |
| Pages | 417-431 |
| Published date | 01 November 2002 |
| Author | Timothy K. Shih,Chuan‐Feng Chiu,Hui‐huang Hsu,Fuhua Lin |
| Subject Matter | Economics,Information & knowledge management,Management science & operations |
An integrated framework for recommendation
systems in e-commerce
Timothy K. Shih
Multimedia and Information Network Lab (MINE Lab), Department of
Computer Science and Information Engineering, Tamkang University, Taipei,
People's Republic of China
Chuan-Feng Chiu
Multimedia and Information Network Lab (MINE Lab), Department of
Computer Science and Information Engineering, Tamkang University, Taipei,
People's Republic of China
Hui-huang Hsu
Department of Computer Science, Chinese Culture University, Taipei,
People's Republic of China
Fuhua Lin
Athabasca University, Athabasca, Edmonton, Alberta, Canada
1. Introduction
Because of the popularity of the Internet,
many traditional social activities have been
changed. In recent years electronic
commerce has become a popular trend on the
Internet. Being without spatial or temporal
constraints are the advantages of the
Internet. These benefits make the business
process more flexible. The applications in
electronic commerce (e-commerce)
environment can be divided into categories
that include business-to-business and
business-to-customer applications. Business-
to-business applications focus on the
enterprise activities. That includes
enterprise information management,
information exchange, and supply chain
management, etc. On the other hand,
business-to-customer applications focus on
the relation of merchants and consumers.
Moukas et al. (1998) proposed the consumer
buying behavior model, which describes the
fundamental stages of buying behavior. The
stages include need identification, product
brokering, merchant brokering, negotiation,
purchase and delivery, and product services
and evaluation. These functions are complex
and huge. Hence, we cannot conclude all
processes. In this paper, we focus on the
business-to-customer application and the
product and merchant brokering stage in the
consumer buying behavior model.
In traditional business behavior we get
recommendations or suggestions from
friends and news channels. But in the
Internet environment, customers buy
products without assistance or suggestions.
Thus the recommendation system becomes
an important application area and an
academic research topic. The
recommendation process can be divided
into online processes and offline processes.
At the online stage we discover the
customer's behavior and transform the
information to useful data for analysis. At
the offline stage we analyze the collected
data and classify customers' behavior.
Merchants can use the application to
recommend items to customers based on
their behavior and to increase the sales.
The first process in the work is to know
consumers' interests and behavior.
Information overload is another critical
issueinthee-commercearea.Itisnot
possible for the users to process huge
amounts of information. Thus
recommendation systems are useful
applications in the area, especially for the
recommending product items in the
e-commerce environment. Furthermore,
the community is a special situation on the
Internet. Not only can we recommend to the
user on a certain site but also to the user on
other sites among the site communities. On
the basis of the recommendation system
andsitecommunities,weproposea
recommendation architecture based on
mobile agent technology.
We have surveyed several related research
areas and explain briefly in section 2. We
describe the system software components in
section 3. In section 4, we propose our
research approach and discuss the
recommendation engine in detail. We then
delineate our implementation method and
mobile agent framework in section 5. Finally,
we draw our conclusion and discuss the
future work in section 6.
The current issue and full text archive of this journal is available
at
http://www.emeraldinsight.com/0263-5577.htm
[ 417 ]
Industrial Management &
Data Systems
102/8 [2002] 417±431
#MCB UP Limited
[ISSN 0263-5577]
[DOI 10.1108/02635570210445853]
Keywords
Electronic commerce,
Recommendations, Internet,
Agents, Computer user groups
Abstract
The Internet has become a popular
medium for information exchange
and knowledge delivery. Several
traditional social activities have
moved to the Internet, such as
distance learning, tele-medical
system and. traditional buying and
selling activities. Online merchants
must know what users want, so
providing recommendation services
is an important strategy. Analyzes
users' on-line behavior and
interests, and recommends to them
new or potential products. The
analysis mechanism is based on the
correlation among customers,
product items, and product
features. An algorithm is developed
to classify users into groups and the
recommendation is based on the
classification. The system can help
merchants to make suitable
business decisions and provide
personalized information to the
customers. A generic mobile agent
framework for e-commerce
applications is proposed. The
aforementioned collaborative
computing architecture for the
recommendation system is based
on the framework.
To continue reading
Request your trial