background image
3.2
Software Agents
19
Classification of Agents
A common classification scheme of intelligent agents is the weak and strong notion
of agency described by Wooldridge and Jennings [1995]. In the weak notion of
agency, agents have their own will (autonomy), they are able to interact with
each other (social ability), they respond to stimulus (reactivity), and they take
initiative (pro-activity).
In the strong notion of agency the weak notions of
agency are preserved, in addition agents can move around (mobility), they are
truthful (veracity), they do what they're told to do (benevolence), and they will
perform in an optimal manner to achieve goals (rationality).
Agents for Web Intelligence
Software Agents or Web Agents have previously been used numerous times in
Web Intelligence applications throughout the last decade, e.g. the agent Letizia
that supports Web-browsing and the ContactFinder agent that answer bulletin
board questions, Mladenic [1999].
Scalable Agent Systems
Agent systems supporting Web Intelligence applications need to be scalable in
order to handle thousands of concurrent agents representing users. Underneath
two different views on scalability requirements for agent systems are described,
the second complements the first by porposing that the dimensions of an agent
system should be adaptive.
Multi-agent systems must be self-building (automatically find the optimal organi-
zation at runtime) and adaptive (change this organizational structure according
to changes in the environment) in order to be scalable, Turner and Jennings
[2001].
Multi-agent systems need to scale in at least 4 dimensions (Omer F. Rana and
K. Stout [2000]):
1. Changes in the number of agents on one platform
2. Changes in the number of agents on a set of connected platforms
3. Changes in the size of the data the agents are operating on
4. Changes in the diversity of the agents

<< - < - > - >>