Expert Systems

Expert Systems

Expert Systems - The 1970s were characterized by the development of expert systems.

Expert systems have two components:

1. An inference engine

2. An experience or knowledge base

The first component controls the application of information contained in the second component. The development and use of expert systems raised questions about the way knowledge is represented.

Improving the ways knowledge is represented so that it is more explicit but, at the same time, more concise characterized the 1980s. The 1990s will see a rapid growth in the use of artificial intelligence in automated assembly and other manufacturing applications as the concept of machine learning evolves.

It is generally accepted that a key characteristic of human intelligence is the ability to learn. For artificial intelligence to reach its potential, the concept of machine learning must be fully developed. Machines learn in one of three ways:

1. Parameter adjustment

2. Concept formation

3. Evolution of structure

Parameter adjustment is the most basic of the three approaches to machine learning. It involves adjusting the values of the parameters of a predetermined representation. It can affect only the values and not the structure of the predetermined representation.

Concept formation involves grouping related objects into categories or groups. Evolution of structure makes use of neurocomputing. This involves the parallel activity of elements that are able to communicate the results of computations among themselves.