Ad Code

Ticker

6/recent/ticker-posts

Knowledge Based Expert System (KBES) | Meaning, Features & Components

Knowledge Based Expert System


What is Knowledge Based Expert System (KBES) ?


One of the principal members of the Al group is Knowledge-Based Expert System (KBES). With the advancement of computing facilities and other resources, the focus has shifted to more demanding tasks (which require intelligence).

Meaning of Knowledge Based Expert System


As society and industry become more knowledge oriented, they rely on different experts for solving problems and decision making. Here, KBES becomes a productive tool as it provides collective knowledge of one or more experts. It acts as an expert on demand, anytime, anywhere and also helps in saving money by getting cheaper expert knowledge and letting users' function at a higher level with consistency.

KBES is a computer-based system which uses and produces knowledge from data, information and knowledge. These systems have the ability to understand the information that is being processed and can take a final decision based on it. This is different from the traditional computer systems that do not have any idea about the data/information which they are processing.

A knowledge based expert system is software or computer based information system which comprises knowledge as data that is processed by computer programs to draw inferences from knowledge base.

Features of Knowledge Based Expert System


The key features of the knowledge-based expert system in MIS are as described below:

1) High-Level Expertise : 
The high-level expertise provided to help in problem solving is one of the most valuable features of an expert system. This expertise is analogous to the best thinking by top experts in the specific field providing solutions which are creative, precise and efficient.

2) Predictive Modelling Power : 
Predictive modelling power is another useful feature of an expert system. It acts as a model of problem solving in the given domain and throws up answers for given problem situations and also reflects the change in them (given new situations explaining the reason for change). This helps the user to assess the potential effect of new facts or data and know about their relationship to the results.

3) Institutional Memory : 
Institutional memory refers to the body of knowledge which defines the ability of an expert system. When people in important position leave an organisation, their expertise gets reserved as institutional memory. This is significant and critical for vital military and government institutions where personnel transfers are frequent.

4) Ability to Provide a Training Facility : 
An expert system is also capable of providing training to important personnel. As expert systems already have the knowledge and reasoning ability, they can train various employees when required.

Components of Knowledge Based Expert System


A knowledge base, the Inference Engine (IE) (a search program) and user interface are included in Knowledge Based Expert System (KBES) :

1) Inference Engine (IE) : 
IE is used to understand the knowledge present in the knowledge base.

2) Knowledge Base : 
Knowledge base is the storehouse of different forms of knowledge.

3) User Interface : 
An appropriate user interface must be present which should have the natural. language processing facility.

In KBES, an empty workspace is available for storage of temporary results and pieces of information (knowledge). Just like an expert is valued for her/his explanation and reasoning capabilities, an expert system's credibility also depends on the same. Human beings can forget knowledge/information which she/he does not use regularly and at the same time learn new things. An essential component of a KBS is simulation of such learning, the degree of which determines the lifespan of a KBS.

A KBS can be manually or automatically updated via a machine. However, the basic frame of a KBS rarely needs to be modified. Figure shows all components of a KBS :

Knowledge Based Expert System

Post a Comment

0 Comments