Knowledge acquisition

From Wikipedia, the free encyclopedia

This article may require cleanup to meet Wikipedia's quality standards. Please improve this article if you can. (May 2008)

This article does not cite any references or sources. (May 2008) Please help improve this article by adding citations to reliable sources. Unverifiable material may be challenged and removed.

The introduction to this article provides insufficient context for those unfamiliar with the subject. Please help improve the article with a good introductory style.

This article or section may contain original research or unverified claims. Please improve the article by adding references. See the talk page for details. (June 2008)

Knowledge acquisition is the transformation of knowledge from the forms in which it exists into forms that can be used in a knowledge based system (KBS).

Several different types of knowledge must be “acquired” by knowledge based systems, particularly procedural knowledge (of what to do or how to do it) and declarative knowledge (of what is and what is not). Procedural knowledge is about rules, methods, procedures etc. while declarative knowledge is about facts, concepts, assertions and so on. Rules and methods are usually specific to particular tasks, while facts and concepts can often be more generic.

It’s usually not possible to transfer a domain’s expert knowledge directly to a KBS because the respective representations of knowledge are too dissimilar. Domain experts often explain their knowledge thought the use of anecdotes and examples, but for a KBS more general principles are required. Experts frequently provide incomplete and even incorrect knowledge, or may not be able to articulate their knowledge at all. Experts may not have the required attitude to communicating their knowledge, or insufficient time or resources to do so properly. Also multiple experts may have significantly differing opinions on what is or is not the right or wrong way to do things.

As a consequence knowledge bases are hard to build and computational representation of knowledge is complex. This problem is referred to as the “knowledge acquisition bottleneck”. Multiple iterations of the acquisition process are often required in order to approach correctness, and for this reason knowledge acquisition is sometimes described as an “evolutionary process”.