Recently, more and more webpages are no longer just a collection of media objects, but allow interaction with the user. The user can buy, bid, vote, subscribe and register electronically. Before a user triggers such an online transaction a decision has to be taken, whether and how a particular transaction is compatible with her current beliefs, desires and intentions. In many cases, the information relevant to a particular decision will be conflicting as well as the decision criteria, so that only a careful analysis of the pros and cons will lead to a rational decision. This tutorial shows how virtual sales agents can help in such a situation.
The current generation of internet agents for eCommerce, that are often called shopbots, pro-duce simple comparisons between product offers that are based on a single criterion: the price of a particular product that the user wants to buy. With the new generation of virtual sales and shopping agents, it will become possible to provide the user with customized, multi-dimensional comparisons of products including the price, service quality, maintenance guarantees, and shipping conditions. In addition, similar products or services that match multiple objectives of the user can be compared and presented in a dialectic way, so that active decision support is provided. This will drastically increase the value of electronic shopping for the consumer.
We will explore new styles of eCommerce presentations by making use of multiple virtual sales agents. The basic idea is to communicate information by means of animated dialogs. For instance, a debate between two characters representing contrary opinions is an effective means of providing an audience with the pros and cons of a particular product.
We introduce the concept of a virtual webpage and discuss the role of high-level ontological an-no-tations for the generation of such third generation autoadaptive webpages. A virtual webpage is generated on the fly as a combination of various media objects from multiple web sites or as transformation of a real webpage. It looks like a real webpage, but is not persistently stored. A virtual webpage integrates generated and retrieved material in a coordinated way. It can be tai-lored to a user profile and adapted to a particular interaction context. It has an underlying representation of the presentation context so that a virtual sales agent can comment, point to, and explain its components. We will show that a simple agent communication language as a collec-tion of speech-act like message types is not enough for the realization of intelligent sales agents. An expressive ontology language like OIL is needed to facilitate the operation of virtual sales agents in the semantic web.
The tutorial will be augmented by numerous videos of deployed systems that DFKI has imple-mented for commercial clients.