Page 317 -

300 Chapter 8

Some companies receive so much e-mail that they have to employ clerical worker
to sift through the fl ood of e-mail, answering basic queries and forwarding others to
specialized workers. Others use intelligent fi ltering software such as GrapeVine for
Lotus, which reads a pre-established knowledge chart to determine who should receive
what mail. Intelligent agent services can supplement but not replace the value of
edited information. As information becomes more available, it becomes more and
more crucial to have strong editors fi lter that information ( Webb 1995 ). There is so
much content out there that the tools that fi lter content are going to be as important
as the content itself ( Wingfi eld 1995 ). As stated by the Rutherford Rogers, “ we are
drowning in information but starved for knowledge ” ( Rogers 1985 ).
An end user, required to constantly direct the management process, is the contrib-
uting factor to information overload. But having agents to do the tasks, such as search-
ing and fi ltering, can ultimately reduce the information overload to a degree. Maes
(1994) describes an electronic mail fi ltering agent called Maxims. Maxims is a type of
learning agent. The program learns to prioritize, delete, forward, sort, and archive mail
messages on behalf of a user. The program monitors the user and uses the actions the
user makes as a lesson on what to do. Depending upon threshold limits that are con-
stantly updated, Maxims will guess what the user will do. Upon surpassing a degree
of certainty, it will start to suggest to the user what to do.
Maes (1994) also describes an example of an Internet news-fi ltering program called
NewT. This program takes as input a stream of Usenet news articles and gives as output
a subset of these articles that is recommended for the user to read. The user gives NewT
examples of articles that would and would not be read, and NewT will then retrieve
articles. The user then gives feedback about the articles, and thus NewT will then be
trained further on which articles to retrieve and which articles not to retrieve. NewT
retrieves words of interest from an article by performing a full-text analysis using the
vector space model for documents. Some additional examples of information fi ltering
agents are shown in table 8.3 .
News agents are designed to create custom newspapers from a huge number of web
newspapers throughout the world. The trend in this fi eld is toward autonomous,
personalized, adaptive, and very smart agents that surf the net, newsgroups, databases,
and so on, and deliver selected information to their users. “ Push ” technology is strictly
connected to news bots development, consisting basically in the delivery of informa-
tion on the web that appears to be initiated by the information server rather than by
the client. Some examples are shown in table 8.4 .
Information overload is a problem of the world today, but intelligent agents help
reduce this problem. Using them to fi lter the oncoming traffi c of the information