In the AI arena, there is a knowledge representation technique called a semantic network. A semantic network is created using a structure consisting of nodes and links. The nodes represent objects, concepts, or situations within a specific domain. The links represent and define relationships between the nodes. Semantic networks are often used to represent the knowledge of human experts in AI applications called inference engines or expert systems.

In 1999 an international standard was developed to describe a mechanism for representing information about the structure of information and organizing it into "topics". These topics have occurrences and associations that represent and define relationships between the topics. Information about the topics can be inferred by examining the associations and occurrences linked to the topic. A collection of these topics and associations is called a topic map.

Even at a high level there is an apparent similarity in the structure of these concepts. This similarity led the author to explore some interesting possibilities:

• Is it possible/reasonable to build a semantic network from a topic map?
• Is it possible/reasonable to store semantic network information in a topic map?
• Would it be possible to design a computer program that identifies the knowledge contained within chunks of text?
• If such a program could be built, would a computer be able to identify and interpret the knowledge found within a collection of documents?

In such a system, a user might be able to query the database for specific information. This system could be used to interpret the knowledge contained within the nodes. The user could begin a browsing session based on a piece of knowledge desired. The user could also request that the system interpret the knowledge in the database without manually browsing through the nodes.

This paper will discuss topic maps and semantic networks and how the two concepts may interrelate. Issues with the topic map standard that make knowledge representation more difficult will be discussed. Also a semantic network system built on topic maps will be presented.