Relevance

Relevance is a key area for AI - in order to process logical deduction more efficiently, we need a notion of relevance, so that when presented with a problem, our machine knows what kind of information it should be looking for in order to answer the question,

Relevance is also a key area for modelling - when we model a system, especially at the beginning of the process, we need to know what are the key items that need to be noted down.

Information has importance, which we need to codify. Google codifies web information importance through the pagerank algorithm.

Importance of information is something that has both an objective stance and a relative stance. Google PageRank records the importance of information in the WWW according to the relative importance of other web sites and the links to that web site.

Thus in the absence of any model of how the information might be used, Google can provide a ranking of information. To this ranking are applied search terms to arrive at the most appropriate knowledge for a particular page.

Relevance is also important from a business system design point of view. In the same way that we want to minimise risk on any activity, we also want to minimise effort, and this means anticipating all knowledge based items beforehand that might affect the development of any knowledge work.

High level relevance work includes the subdivision of work into components so that each work item can be manged independently.

Any project should include a relevance phase as part of the risk management in which knowledge is classified into relevance categories or relevance hiearchies.