Expert System and Knowledge Based Systems
An Expert system is a type of application program that makes decisions or solves problems in a particular field by using knowledge and analytical rules defined by experts in that field. A Knowledge-Based System is a system that uses the knowledge provided with to solve problems in a specific domain. Books and manuals have a tremendous amount of knowledge but a human has to read and interpret the knowledge for it to be used. Taking this into consideration it can be said that a human cannot perform a complex problem because of the different disadvantages he/she might be facing. The disadvantage of being unreliable, speed, and not enough memory capacity would cause a human to make a mistake or be inefficient. For this case Humans can design Expert systems providing the human intelligence and knowledge of solving a specified problem.
These Expert systems and knowledge based systems have underlying rules which they are bound to helping them in solving complex problems. Enough information must be fed to the expert system to make it able to accomplish the different tasks the human would want it to do. Knowledge-based systems collect the small fragments of human know-how into a knowledge-base which is used to reason through a problem, using the knowledge that is appropriate. The ability of these expert systems to explain the reasoning process through which they carry is a feature which they can not do.
The Below diagram represents how knowledge travels from state to state to then end user interface of an expert system
The knowledge resides in the hard disk, processed in the inference engine and later, the user interface can be used to produce outputs of the solutions.
Purpose of expert systems and knowledge based systems
Expert systems are meant to solve real problems which normally would require a specialized human expert (such as a doctor). Building an expert system therefore first involves extracting the relevant knowledge from the human expert. There are several purposes to emphasis on knowledge-based methods rather than other formal representations and associated analytic methods.
The main goal of the expert system research is to get very quick answers for the technicians and the markers who use this kind of expertise. That answers are not available always at the right place and right time because there is no enough expertise to looking for by them. . Portable with computers loaded with in-depth knowledge of specific subjects can bring decade’s worth of knowledge to a problem. Many of the supervisors and managers use the same system to help them with situation assessment and long-range planning. During these times, many small systems exist that bring a narrow slice of (in-depth knowledge) to specific problems, and that prove to us that the broader goal is achievable.
Expert systems (the knowledge based systems) of AI (artificial intelligence) have enhanced productivity in business, science, engineering, and the military with some advances in the last decade. Expert systems today can be chosen from dozens of commercial software packages which is easy to use interfaces.
AI researches provide even better applications of expert systems because each new deployment of an expert system yields valuable data for what works in what context.
Knowledge-based system versus expert system
Knowledge-based system is a more general than the expert system. As the below diagram will illustrate the difference, knowledge based are a verse domain than an expert system. Expert systems are a specified domain of a knowledge based as expert system uses the representation of knowledge to solve problems.
As simple as it looks, expert systems are well known to solve complex problems. Computers are good in representing numbers, words and even maps but the greatest difficulty it faces is representing knowledge.
Knowledge based construction
Knowledge can be distinguished from mere facts by the way it is used in decision processes. There are two approaches leading to successful knowledge based system. The approaches can be by obtaining expert knowledge on a specific problem and breaking the facts into rules which can be applied to solve a problem. The other method is through learning through experience. As a system functions, whatever could be seen as a malfunction or an extra requirement is filled in. These seem simple methods to address, but they are indeed obtaining through much effort.
Types of Expert systems and knowledge-based systems
a) Rule base
a) Rule Based
“Instead of representing knowledge in a relatively declarative, static way (as a bunch of things that are true), rule-based system represent knowledge in terms of a bunch of rules that tell you what you should do or what you could conclude in different situations. A rule-based system consists of a bunch of IF-THEN rules, a bunch of facts, and some interpreter controlling the application of the rules, given the facts. ”
In programming, object-oriented is a computer programming paradigm. Many programming languages support object-oriented programming. Many programming frameworks, like the Java platform and the .NET Framework, are built on object-oriented principles. Object-oriented programming is often abbreviated as OOP.”
Information verses Knowledge based system
Information can be termed as raw data waiting to be processed to attain a goal. Knowledge based system on the other hand is the engine which uses such information processes it in to rules and facts which can be used in archiving a specified goal. The figure below illustrates clearly the Information verses Knowledge based system.
The DENDRAL is one of the earliest systems to cover the domain-specific knowledge in problem solving. It was developed at Stanford in the late 1960’s by Lindsay. The DENDRAL was designed to be able to recognize the structure of organic molecules from their chemical molecules. The Number of molecules is very large to be able to refer to books often. The DENDRAL uses the heuristic knowledge developed by chemical experts to clarify the problem from the structure f the molecule. DENDRAL was a success in only a few trials and was marketed all over the world.
MYCIN was developed by Buchanan and Shortliff in the mid-1970 in Stanford. MYCIN can be also described as a system which acts like a doctor in a hospital. It uses expert medical knowledge to diagnose and prescribe treatment for spinal meningitis and bacterial infections of the blood. MYCIN is the first program which had the capability to reason with uncertain or incomplete information and provided clear and logical explanations of its reasoning.
Prominent expert systems
- Dendral:- analyze mass spectra
- Dipmeter:- Advisor analysis of data gathered during oil exploration
- Mycin:- diagnose infectious blood diseases and recommend antibiotics (by
- CADUCEUS:- (expert system) blood-borne infectious bacteria
- R1 (expert system)/ order processing
- CLIPS:- programming language
- Prolog:- programming language(logic)
- Jess:- CLIPS using Java with more features
- ART:- programming language
Types of problems solved by expert systems
Organizations with highly experienced expertise which the knowledge cannot easily be transferred to other members would value expert system the most. The expert system can be design to carry the intelligence and information found in the experts knowledge and provide such kinds of knowledge for other members of the organization for problem solving purposes.
Most of the problems which require expert system might seem easy to be solved by a professional. Generally expert systems are used for problems for which there is no single “correct” solution which can be encoded in a predictable algorithm. One would not write an expert system to find shortest paths through graphs, or sort data, as there are simply easier ways to do these tasks.
Simple systems use simple true/false logic to evaluate their data, but more sophisticated systems are capable of performing at least some evaluation by taking into account real-world uncertainties.
Taking an Example in predicting the weather forecast might seem a simple task to do. The probability your answer stays correct would be very small compared to the sophisticated systems which take every data into consideration and provide a higher chance of success.
- The ES is a repository of valuable information that might otherwise be lost and inaccessible to the firm creating the system and even useful to state them explicitly.
- The ES can be indispensable when human expertise is not accessible. This could be critical in disciplines such as medicine and in remote areas.
- ES could be more efficient and cost effective than human systems, and will become increasingly so as wages of human professionals rise.
- The ES could be better than local and even national human expert if the expertise of world-renowned experts is captures within the knowledge-base of the system.
- An ES that is predictive can be particularly valuable when the predictions are generated fast and tirelessly.
- An ES can be used for training future human experts. One such system can be duplicated, at very little cost, to yield as many copies as are required.
- An ES can be used also:
- To store and be able to manipulate important levels of information
- To be able to provide consistent answers for repetitive decisions, processes and tasks
- Reduce employee training costs
- Centralize the decision making process
- Create efficiencies and reduce time needed to solve problems
- Combine multiple human expert intelligences
- Reduce the amount of human errors
- Review transactions that human experts may overlook
- An ES cannot reason on the basis of human ‘gut feeling’, of intuition, or even of common sense, because these modes of reasoning are not easily represent able as a knowledge-base of rules and facts.
- An ES is confined to a restricted domain of expertise; it cannot easily integrate expertise from other domains, nor an it generalize reliably.
- Many of the conceptually complex and tough problems in business, industry, and society do not appear to be applicable to current ES technology.
- Current ES cannot reason reliably from theories or from analysis.
- The knowledge in an ES is highly dependent upon the human expert expressing and articulating knowledge in the form that can be used in a knowledge-base.
- The lack of human common sense needed in some decision makings
- The creative responses human experts can respond to in unusual circumstances
- Domain experts not always being able to explain their logic and reasoning
- The challenges of automating complex processes
- The lack of flexibility and ability to adapt to changing environments
- Not being able to recognize when no answer is available
In general, expert system and knowledge systems can solve complex problems, this cannot be seen as a disadvantages. Keeping in mind, Expert system can cause lack of employment in a society where it is implemented. Expert and knowledge based system are to be analyzed well before being implemented if there are critical drawbacks then it is better off. I have a strong belief that the government should keep track of the number of such systems in the country as the more the systems are created, the tougher employment gets, due to this, life might become easier to live in, but that’s not always a good thing.
No related posts.