The Knownet Intelligent Agents Page

This page will contain introductory material about agent technology and intelligent agent applications for Knowledge Management, it will host Knownet specific discussions and shared information, and shall support that the Knownet agent discussions get more detailed in the workpackages 2 (agent awareness within the requirements engineering process), 3 (agent specification within tool design), and 4 (tool development). It should be as specific for the Knownet purpose as possible, avoiding the flood of less valuable information about all aspects of all facets of the agent concept which can usually be found at the general agent resource pages whilst pointing to these repositories for further reference. I hope you will contribute to the page and discuss things.

Contents:



Updates:

Until we have agreed upon an easy and safe way of cooperation via Knowledger or whatever tool, please send me your suggestions and modifications by mail, I will include them in this page.

Apologies:

Accept my apologies that this page has not been created weeks before. Basically, it was intended as the first input for "agent awareness creation" within the requirements work package, and I planned parts of the content already during our first meeting in October '98 in Athens. However, what made me very reluctant in this case was the fact that the agent community and scope is much too broad, diverse, and unstructured for having really useful input for our project at first hand. Furthermore, I was not sure how appropriate the agent notion would be in our case and how it would fit into our Knownet system conception. But since waiting too long for an optimal solution does not help anyone, I now start with the "protptyping" approach, just putting something on the web to start collaboration and let the solutions evolve ;-)


 Introductory Agent Papers

For a definition of Intelligent Agents we quote from a White Paper of Reticular Systems Inc.:
 
The concept of an intelligent software agent has captured the popular imagination. People like the idea of delegating complex tasks to software agents. These agents can make airline reservations, order new books from an online store, find out about the latest song from a favorite musician, or monitor stock portfolios. Software agents can roam the Internet finding information for us. Sophisticated software agents can negotiate the purchase of raw materials for a factory, schedule factory production, negotiate delivery schedules with a customer's software agent, or automate the billing process. (...)

Intelligent software agents are a new class of software that act on behalf of the user to find and filter information, negotiate for services, easily automate complex tasks, or collaborate with other software agents to solve complex problems. Software agents are a powerful abstraction for visualizing and structuring complex software. (...)

The central idea underlying software agents is that of delegation. The owner or user of a software agent delegates a task to the agent and the agent autonomously performs that task on behalf of the user. The agent must be able to communicate with the user to receive its instructions and provide the user with the results of its activities. Finally, an agent must be able to monitor the state of its own execution environment and make the decisions necessary for it to carry out its delegated tasks.

There are two approaches to building agent-based systems: the developer can utilize a single stand-alone agent or implement a multi-agent system. A stand-alone agent communicates only with the user and provides all of the functionality required to implement an agent-based program. Multi-agent systems are computational systems in which several agents cooperate to achieve some task that would otherwise be difficult or impossible for a single agent to achieve. We term these multi-agent systems agencies. Agents within an agency communicate, cooperate, and negotiate with each other to find a solution to a particular problem.

Many authors have many definitions.  Some often mentioned characteristics of agents are:

The following three introductory papers give a comprehensive view on the agent research area. Most of them are taken from the University of Maryland AgentWeb resource page.
 


Intelligent Agents for Knowledge Management




 

Agent technology: systems, tools, languages, standards, ...


Source:

http://www.cs.umbc.edu/agents/technology/
 

Source: http://www.operant.com/agents.htmlich
 


Agents on, by and for the Web

http://www.cs.umbc.edu/agents/technology/misc.shtml

Commercial

The number of companies providing some sort of agent product has exploded in the last couple of years. Some companies, such as Firefly and NetBot, were founded by academic researchers and now offer a commercial product. Other companies offer consulting. Still other companies have on-going research efforts which might eventually lead to products.
Andersen Consulting
BargainFinder, LifestyleFinder, InfoFinder, etc.
A T & T
A number of papers by Bart Selman and Henry Kautz.
Autonomy Corporation
Agentware i3. Uses "concept clustering," fuzzy logic, and a type of neural network they call "Dynamic Reasoning Engine."
Crystaliz, Inc.
LogicWare agent server.
Empirical Media
WiseWire Service
They offer a personalized information service, derived from NewsWeeder project. They also offer a service bureau-like filtering and ad targeting service, as well as licensing their agent technology for other desktop applications.
Enterprise Integration Technologies
A Verifone company, involved in electronic commerce, and in CommerceNet.
The SmartProcurement System
Extempo Systems
Synthetic characters that can add improvisational behavior to products.
Firefly Network, Inc.
Formerly Agents, Inc. Pattie Maes and her students from MIT Media Lab were among the founders.
Firefly
An on-line community. Matches your music preferences via vector search to others like you, and makes recommendations based on what they have liked.
The Red Herring interview of Pattie Maes and Nick Grouf.
First Floor Software
Agents incorporated into document delivery systems from the Web to your desktop.
General Magic
One of the most established commercial players in the agents field, General Magic has been trying to reposition itself and focus on a new network service. The Telescript architecture has been around a long time, but check out new developments around Odyssey.
Odyssey
A new mobile agent technology, 100% Java. The class libraries are extensible allowing you to integrate their agent technology with your application.
mobilis April 1996: Jim White Interview
Gentia Software plc
Intelligent agent software explained
IBM
IBM has a number of different products and efforts underway with intelligent agents, (although the future of "aglets" at IBM is unclear now that the Chief Architect of aglets has left for General Magic).
IBM Intelligent Agent Center of Competence
IBM Aglets Workbench - Home Page
IBM Agent Building Environment
IBM Web Browser Intelligence
MCC
Microelectronics and Computer Technology Corporation, based in Austin, Texas.
InfoSleuth Project Index
MCC project, follow-on to Carnot. Has demonstrations, but they're password protected.
Carnot Project Details
Netbot, Inc.
Company formed by Etzioni and Weld, to commercialize their softbot technology. The product is called Jambo.
Oz, Inc.
Virtual Reality agents "with attitude." The web page hints at intelligent technologies such as neural networks, genetic algorithms, etc. but it never comes right out and says what they use.
Verity
Agent Server and Agent Server Toolkit are of interest as search agent products.
Xerox PARC
Dynamics of Computation Area, B.A. Huberman and T. Hogg. There's some interesting research on multi-agent systems, and some papers on "controlling smart matter."
Zuno
Owned by Mitsubishi Electric. They claim to "develop leading-edge agent-based software components and tools." Beyond that, their web site leaves you wondering what they do.
 


Points of application for agent technology in Knownet