Wave Issue 9039 X/XX/99 Article 1-03

***NewsMaps.com Shows Visualization Technology Topographical Maps
for News
(April 6)

Advanced visualization technology is being used on NewsMap to
show thousands of news articles or discussion postings at once,
all visually organized by content rather than categories. Each
NewsMap typically includes links to several hundred articles from
a variety of major news sources on the Internet.

The concept behind a NewsMap is simple: the greater the
similarity between any two documents, the closer together they
appear on the map. Peaks appear on the map where there is a high
concentration of documents about the same topic. The distance
between peaks shows how closely the topics are related. For
example, a map of news articles about technology might have peaks
labeled "Internet" and "Online Shopping" in close proximity, but
relatively distant from a peak labeled "Cell Phones." Individual
documents appear as dots on the map, and users can zoom in to
reveal greater detail. For any region on the map, a click of the
mouse button pops up a list of documents with related content, a
result of NewsMaps' organization. Pointing to any document title
displays a short summary. A mouse click links directly to the
original document at its source on the web. NewsMaps.com also
provides extensive search capabilities for documents within a
map.

Users can enter keywords or select from a list of topics
discovered during the organization process. Unlike conventional
search engines, NewsMaps does not return multiple pages of search
results. All matching documents are ranked and highlighted on the
map itself, each located according to the information it
contains. With very little reading, users can rapidly isolate the
information they need. NewsMaps.com is a free site. A small Java
applet provides interactive control and is downloaded when the
first map is selected. Access requires Netscape Navigator or
Microsoft Internet Explorer, version 3 or higher. Version 4
browsers are recommended.

Cartia develops Relational Topic Mapping (RTM) software, a
technology that automatically extracts the content of
unstructured text and organizes it onto interactive maps of
information. The company's commercial product, known as
ThemeScape, automatically harvests massive quantities of
documents and webpages, extracts the primary themes and concepts,
and creates a relational topic map of the entire information
landscape.

The publishing system cost approximately $20,000.

www.cartia.com