Business Systems and Knowledge Modelling Lab

Research Project

Context-Aware E-Commerce

Research Goal
A human-centered (as against technology-centered) approach to e-Commerce is developed. We also determine how semi-structured, XML-based data model can be exploited to create a context-dependent management of information.

Funding
Grants worth $110,000 received

Problem Space
Electronic Commerce (EC) can be broadly seen as the application of information technology and telecommunications to create virtual trading networks where goods and services are sold and purchased.

A number of software architectures have been proposed for supporting such trading networks in the past few years; in Hands et al. (1998) some of them are presented, addressing issues such as sales, ordering and delivery of products in the framework of the global Internet. Besides greatly increasing the efficiency and the effectiveness of traditional commerce, EC techniques stimulated interest in new distribution techniques for multimedia digital content.

Interestingly, experience has shown that transition between physical products and digital ones is usually gradual rather than abrupt. E-commerce transactions involving so-called smart products often require transfer or presentation of some sort of digital content to the customer in addition to the delivery of a conventional product or service. Digital content is conceptually very different from traditional physical products, being characterized by an increased interactivity and the possibility of multiple uses. Also, digital content is seldom autonomous; rather, it needs to be executed (or displayed) using suitable devices

Taking another perspective, traditional supply chains were based on stable, long-lasting business relationships, while EC trading networks allow and indeed encourage a higher rate of change in commerce relations, causing customers to frequently change suppliers on the basis of short-term business opportunities. It is widely acknowledged that this ever-changing business environment requires that content offered on the Internet be highly adaptive, even more so in comparison with other communication channels such as cable television. New user interfaces and devices keep emerging, the diversity of users is increasing, machines are acting more and more on behalf of humans, and Internet activities are becoming possible for a wider range of business, leisure, education, and research activities.

To achieve the high flexibility that this scenario requires, management of Internet content must be done at a much finer detail of granularity than traditional Web-based systems can provide. For example, content needs to be broken down into tagged information items, equipped with metadata providing enough contextual information. However, we shall not commit to the Semantic Web approach nor to a specific metadata format; rather we research in some detail how XML metadata should be used to ensure e-commerce systems are human-centered. We outline a human-centered (as against technology-centered) approach to e-business.

Team Members
Rajiv Khosla, Ernesto Damiani, Somkiat Kitjongthawonkul, Qiubang Li

Publications
[1] E. Damiani, R. Khosla, "Granulating XML information," in Archives of
Control Sciences, Vol. 12, No. 4, pp. 411-31, December 2002

[2] R. Khosla and S. Kitjongthawonkul, 'A Human-Centered Agent-Based Architecture
for Electronic Brokerag in e-Commerce', in International Journal of Soft Computing, Springer-Verlag, vol 5 no. 3, 2001

[3] R. Khosla and E. Damiani, Human-Centered Agent Oriented Architecture for
Electronic Brokerage Systems, in International ACM symposium on Applied Computing, Washington, pp. 243-249, February, 1999

[4] Q. Li, and R. Khosla. “Intelligent Agent Based Framework for Mining Customer
Buying Habits in E-Commerce (application in e-banking), in 4th International Conf.
on Enterprise Info. Sys, Spain, April 2002