Introduction


One potential application for agent technology is in the area of mobile commerce (m-commerce) (Nwana & Ndumu, 1996; Aylett et al., 1998). According to a study done by Frost and Sullivan, [1] it was projected that electronic commerce (e-commerce) conducted via mobile devices such as cellular phones and Personal Digital Assistants (PDAs) will become a whopping $25 billion market worldwide by 2006. Some of the driving factors behind m-commerce were attributed to the compactness and high penetration rate of these mobile devices.

However, despite all the hype and promises about m-commerce, several main issues will have to be resolved before agent technology can be fully adopted into any m-commerce system (Nwana & Ndumu, 1997; Morris & Dickinson, 2001). Clumsy user interfaces, cumbersome application, low speeds, flaky connections, and expensive services soured many who tried m-commerce. Security and privacy concerns also dampened enthusiasm for m-commerce.

Taking these concerns into account, it seems like good old e-commerce will remain as the preferred choice for online transactions for many years to come. Customers will only use wireless mobile devices to access the Internet if they have a good reason to do so. Therefore, in order to entice customers to participate in m-commerce, the developers will have to offer something that is unique and which no self-respecting consumer can live without. One of the potential "killer" applications for m-commerce could be an intelligent program that is able to search and retrieve a personalized set of products from the Internet for its user.

Currently, when a user wants to search for a particular product on the Internet, what he will normally do is use popular search engines, such as Altavista, [2] and enter keywords that describe the product. These search engines will process these keywords and generate a large number of links for the user to visit.

Although these are the more common methods of searching for information on the Internet, they may not be the best or the most efficient. Neither the search engine nor the Web site know the preference of the user and, hence, might provide information that is irrelevant to the user. For example, if the user wants to search for information about "mobile agents," the search engine could return links to "insurance agents" instead.

In agent-based m-commerce, agents act on behalf of their users by carrying out delegated tasks automatically. Currently, there is no single agent that can perform all the tasks meted out by the user effectively. Like humans, specialized agents are required that are able to work in a specific type of environment. A product-brokering agent seems to be a potential solution for this scenario. The agent will search for the products in the background, with minimal user intervention, thereby allowing the user to concentrate on other tasks. It could be programmed with the user's preferences in mind and filter out irrelevant products automatically. The agent could also detect shifts in the user's interest and adjust accordingly to suit the user.

Described in this chapter is the design of an intelligent ontology-based product-brokering agent capable of providing a personalized service for its user. It does this through user profiling (Soltysiak & Crabtree, 1998a). Such agents are able to learn user preferences over time and recommend products that might interest the user. This technique has been used successfully for certain types of agent tasks, typically those that are information intensive and often involve the World Wide Web.

[1]http://www.infoworld.com/articles/hn/xml/02/03/22/020322hnmcommerce.xml.

[2]http://www.altavista.com.




Wireless Communications and Mobile Commerce
Wireless Communications and Mobile Commerce
ISBN: 1591402123
EAN: 2147483647
Year: 2004
Pages: 139

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net