Both in Computer Science and Artificial Intelligence the term 'Intelligent Agents' are addressed. Today's discussion
will be held within computer science area.
1. Intelligent Agents in Computer Science
In computer science, intelligent agents, also known intelligent software agents or software agents, refer to software
that assists users and will act on their behalf. "Intelligent" in this context is often taken to imply the ability to adapt
2. Types of Intelligent Agents
The document, Third Canadian Edition of "Management Information Systems for the Information Age",
suggests that there are only four essential types of intelligent agents, which are:
2.1 Buyer Agents
Buyer agents are much like search engines specialized in searching goods being sold on-line. They travel around
the network (i.e. The Internet) collecting products and price information from different shopping sites, index the
discovered data and stores in their databases. Once the buyer agents have learnt the products users wish to
purchase, they will generate a list of price and websites' information for users to compare and make decisions. One
example is shopping.com (http://www.shopping.com). The website uses a search engine to help user locate the
product they want to buy and then hand over the product information, user's geographical location (according to
users' IP address) to its buy agent, then a comparison list will be generated and displayed.
2.2 User Agents
User agents work like personal servants on users' behalf. They are 'trained' to learn users preference and to fulfill
such needs. One example is iGoogle (http://www.google.com/igoogle). The user agent of this web site work with a
user profile that records a user's preference. A user can select items, like news from certain newspaper websites,
weather forecast, current time, mail box, games, videos and etc., place selected items in different positions on the
web page and a picture theme of the web page, which are all recorded into its profile. Once the user agent have
learnt a user's needs, it will compose all preferred information into a web page when the user logs in.
2.3 Monitoring-and-surveillance Agents
These agents, also known as "predictive agents", are intelligent agents that observe and report on equipment. For
example, NASA's Jet Propulsion Laboratory has an agent that monitors inventory, planning, and scheduling
equipment ordering to keep costs down, as well as food storage facilities. These agents usually monitor complex
computer networks that can keep track of the configuration of each computer connected to the network. Another
example is an application, which is developed in Chinese Academy of Science, Institute of Automation, uses
surveillance cameras to monitor and identifies suspicious moving objects. The application keeps capturing images
and analyzing the captured images to determine if there are any specious objects enters the monitored space.
2.4 Data Mining Agents
A data mining agent operates in a data warehouse discovering information. A 'data warehouse' brings together
information from lots of different sources. 'Data mining' is the process of looking through the data warehouse to
find information that you can use to take action, such as ways to increase sales or keep customers who are
considering defecting. 'Classification' is one of the most common types of data mining, which finds patterns in
information and categorizes them into different classes. Data mining agents can also detect major shifts in trends or
a key indicator and can detect the presence of new information and alert you to it.
3. A Question to Ask
After going through the four different types of intelligent agents mentioned above, a question is raised - are they