What is an Expert System?

Expert system is a crucial subset of artificial intelligence (AI) that is designed to solve complex
problems and to provide decision-making ability like a human expert. In simple terms, An
Expert System is an AI program that mimics human decision-making to solve complex
problems in a specific field. It uses knowledge and rules to provide expert-level solutions. It
performs this by extracting knowledge from its knowledge base and apply some reasoning
according to the user queries.

The performance of an expert system is based on the its knowledge stored in its knowledge
base. The more knowledge stored in the Knowledge Base, the more that expert system
improves its performance. One of the common examples of an Expert System is a suggestion
of spelling errors while typing in the Google search box.

Below are some popular examples of the Expert System:

DENDRAL: It was an artificial intelligence project that was made as a chemical analysis
expert system. It was used in organic chemistry to detect unknown organic molecules
with the help of their mass spectra and knowledge base of chemistry.

MYCIN: It was one of the earliest backward chaining expert systems that was designed
to find the bacteria causing infections like bacteraemia and meningitis. It was also used
for the recommendation of antibiotics and the diagnosis of blood clotting diseases.

Characteristics of Expert System

High Performance: The expert system provides high performance for solving any type
of complex problem of a specific domain with high efficiency and accuracy.

Understandable: It responds in a way that can be easily understandable by the user. It

can take input in human language and provides the output in the same way.

Problem explanation: It can provide a detailed description of a given problem.

Reliable: It is much reliable for generating an efficient and accurate output.

Highly responsive: ES provides the result for any complex query within a very short
period of time.

Structure of Expert System

User Interface

Inference Engine

Knowledge Base
1. User Interface
The user interface is like the bridge that allows users to interact with the expert system. With
the help of a user interface, the expert system interacts with the user, takes queries as an
input in a readable format, and passes it to the inference engine. After getting the response
from the inference engine, it displays the output to the user. In other words, it is an interface
that helps a non-expert user to communicate with the expert system to find a solution.

2. Inference Engine(Rules of Engine)

The inference engine is known as the brain of the expert system as it is the main
processing unit of the system. It applies inference rules to the knowledge base to derive
a conclusion or deduce new information. It helps in deriving an error-free solution of
queries asked by the user.

With the help of an inference engine, the system extracts the knowledge from the
knowledge base.

There are two types of inference engine:

Deterministic Inference engine: The conclusions drawn from this type of inference
engine are assumed to be true. It is based on facts and rules.

Probabilistic Inference engine: This type of inference engine contains uncertainty in

conclusions, and based on the probability.

Inference engine uses the below modes to derive the solutions:

Forward Chaining: It starts from the known facts and rules, and applies the inference
rules to add their conclusion to the known facts.

Backward Chaining: It is a backward reasoning method that starts from the goal and
works backward to prove the known facts.

3. Knowledge Base
The knowledgebase is a type of storage that stores knowledge acquired from the
different experts of the particular domain. It is considered as big storage of knowledge.
The more the knowledge base, the more precise will be the Expert System.

It is similar to a database that contains information and rules of a particular domain or

subject.

One can also view the knowledge base as collections of objects and their attributes.
Such as a Lion is an object and its attributes are it is a mammal, it is not a domestic
animal, etc.

Components of Knowledge Base

 Factual Knowledge: The knowledge which is based on facts and knows by knowledge
engineers comes under factual knowledge.

Heuristic Knowledge: This knowledge is based on practice, the ability to guess,

evaluation, and experiences.

Knowledge Representation: It is used to formalize the knowledge stored in the knowledge

base using the If-else rules.

Knowledge Acquisitions: It is the process of extracting, organizing, and structuring the

domain knowledge, specifying the rules to acquire the knowledge from various experts, and
store that knowledge into the knowledge base.

Types of Expert Systems

1. Rule-Based Expert Systems : One of the most common types is Rule-Based Expert
Systems, which rely on if-then rules to process information and make decisions. These
rules are typically crafted by domain experts and serve as the system’s reasoning
mechanism. A well-known example is MYCIN, an early medical diagnosis system that
identified bacterial infections.
2. Frame-Based Expert Systems : Another category is Frame-Based Expert Systems, which
organize knowledge using frames, similar to objects in programming. These frames store
attributes and values related to specific concepts, making them useful in natural
language processing and other knowledge representation tasks.
3. Fuzzy Logic Systems : For situations involving uncertainty and imprecision, fuzzy logic
Systems come into play. These systems don’t operate on strict true/false values but
instead allow for degrees of truth. Fuzzy control systems, commonly used in household
appliances like washing machines and air conditioners, leverage this approach to
optimize performance based on variable input conditions.
4. Neural Network-Based Expert Systems : Integrate artificial neural networks to learn
patterns from data and improve decision-making. These systems are widely used in
applications like image recognition and speech processing, where traditional rule-based
approaches might struggle.
5. Neuro-Fuzzy Expert Systems : A more advanced hybrid approach is Neuro-Fuzzy Expert
Systems, which merge the learning capabilities of neural networks with the uncertainty-
handling strengths of fuzzy logic. These systems are particularly useful in financial
forecasting and automated control systems, where both structured learning and flexible
reasoning are necessary.

Development of Expert System

1. Problem Identification: The first step in developing an expert system is to clearly define
the problem it will solve. The domain of expertise is identified, such as medical diagnosis,
financial analysis, or engineering design. The complexity of the problem is analyzed to
check whether it requires human-like intelligence. Additionally, the availability of domain
 experts or research materials is considered, as an expert system can only function
effectively if enough reliable knowledge is available.
2. Knowledge Acquisition: Once the problem is identified, the next step is to gather
knowledge from various sources. Knowledge can be collected from domain experts,
books, research papers, and case studies. The collected knowledge consists of facts, rules,
and heuristics (experience-based techniques used by experts). The quality of an expert
system heavily depends on how well the knowledge is acquired and structured. If the
information is incomplete or inaccurate, the system may give incorrect outputs.
3. Knowledge Representation: After acquiring knowledge, it must be stored in a
structured manner so that the system can process it effectively. Different techniques are
used for knowledge representation:
Rule-based representation: Knowledge is stored as IF-THEN rules. For example, "IF
a patient has a high fever, THEN suggest a blood test."
Semantic networks: Concepts are represented as nodes, and relationships between
them are shown as links.
Frames: Knowledge is stored in structured objects with attributes and values.
Fuzzy logic: Used when the information is uncertain or imprecise, allowing the
system to handle approximate reasoning.
4. Inference Engine Development: The inference engine is the core reasoning mechanism
of an expert system. It applies logical rules to the stored knowledge and makes decisions.
There are two main reasoning approaches:
Forward Chaining: The system starts with known facts and applies rules to reach
conclusions. This method is commonly used in diagnostic systems, such as medical
expert systems.
Backward Chaining: The system starts with a goal and works backward to find
supporting facts. This is often used in troubleshooting systems, where the system
tries to confirm whether a particular condition is true by checking available evidence.
5. User Interface Design: The user interface allows users to interact with the expert system.
A well-designed interface ensures that users can easily enter queries and receive
explanations for the system’s conclusions. In some cases, expert systems also include an
explanation facility that justifies how a particular decision was reached. This improves
user trust and helps them understand the reasoning behind the system’s
recommendations.
6. Testing and Validation: Before deploying the expert system, it must be tested to ensure
accuracy and reliability. The system is tested using real-world cases and compared with
expert decisions. If errors or inconsistencies are found, modifications are made to
improve accuracy. This phase is crucial because an expert system must be highly reliable
before it can be used in real-world scenarios.
7. Deployment and Maintenance: After successful testing, the expert system is deployed
for real-world use. However, deployment is not the final step. Regular maintenance is
necessary to keep the system updated with new knowledge. As new discoveries and
techniques emerge, the expert system must be modified to incorporate them. If not
 maintained properly, the system may become outdated and provide incorrect
recommendations.

Advantages/Benefits of Expert System

They can be used for risky places where the human presence is not safe.

Error possibilities are less if the Knowledge Base contains correct knowledge.

The performance of these systems remains steady as it is not affected by emotions,

tension, or fatigue.

They provide a very high speed to respond to a particular query.

Cost control : Expert systems are relatively inexpensive compared to the cost of
employing human experts. They can help reach decisions more efficiently, which saves
time and cuts costs.

Permanence : Human experts eventually leave their role, and a lot of specific knowledge
may go with them. Knowledge-based systems provide a permanent repository for
knowledge and information.

Accuracy : Expert systems are not prone to human error or emotional influence. They
make decisions based on defined rules and facts.

Availability: They are available 24/7 and can handle multiple queries simultaneously,
providing timely assistance and support.

Cost-Effectiveness: By automating expert-level decision-making, organizations can save

on the costs associated with hiring and training human experts.

Limitations of Expert System

The response of the expert system may get wrong if the knowledge base contains the
wrong information.

Like a human being, it cannot produce a creative output for different scenarios.

Its maintenance and development costs are very high.

Knowledge acquisition for designing is much difficult.

For each domain, we require a specific Expert System, which is one of the big limitations.

It cannot learn from itself and hence requires manual updates.

Applications of Expert System

 In designing and manufacturing domain
It can be broadly used for designing and manufacturing physical devices such as camera
lenses and automobiles.

In the knowledge domain

These systems are primarily used for publishing the relevant knowledge to the users. The
two popular ES used for this domain is an advisor and a tax advisor.

In the finance domain

In the finance industries, it is used to detect any type of possible fraud, suspicious
activity, and advise bankers that if they should provide loans for business or not.

Planning and Scheduling

The expert systems can also be used for planning and scheduling some particular tasks
for achieving the goal of that task.

Human Expert vs. Expert System

S. No. Human Expert Expert System

1. Problem-Solving Uses experience, intuition, and Solves problems using

Approach practical knowledge predefined rules and logical
(heuristics) to solve problems reasoning in a specific
in a specific field. domain.

2. Knowledge Storage Knowledge is stored in the Knowledge is stored

human brain and applied separately from processing,
naturally. making it easier to update.

3. Explanation Ability Can explain reasoning and Can trace rules used in
provide details on how a problem-solving and explain
conclusion was reached. how and why a conclusion
was reached.

4. Handling Uncertainty Can make decisions even with Can handle some uncertainty
incomplete, uncertain, or but relies on predefined rules.
unclear information.

5. Improvement Over Time Improves through learning, Improves by adding new rules
experience, and practice over or modifying existing ones in
years. the knowledge base.

6. Availability Available only during working Available anytime and can be

hours and may not always be used anywhere without
 accessible. restrictions.

7. Time Taken for Problem- Time varies depending on the Solves problems quickly and
Solving complexity of the problem. takes very little time.

8. Replacement Possibility Cannot be replaced due to Can be replaced with an

adaptability, experience, and improved system if needed.
ability to handle new
situations.

Shells of Expert System

An Expert System Shell is like a basic framework of an expert system, but without any
knowledge. To create a complete expert system, the user just needs to add knowledge in the
form of rules and provide the necessary data.

It provides a simple software environment that helps developers build expert systems without
starting from scratch. The shell already includes the core logic needed for an expert system.

Using an expert system shell makes development easier and faster, but it also has limitations.
Developers must work within the fixed structure of the shell, meaning they cannot freely
choose how to represent knowledge, make inferences, or manage uncertainty. Because of
this, expert systems built with shells can solve problems efficiently but may lack advanced
customization. Today, many expert system shells are available in the market, each with
different features and capabilities.

Components of Expert System Shells

1. Knowledge Base

Stores important facts, rules, and relationships related to a specific field.

Acts as the brain of the expert system, containing all the expertise needed to make
decisions.
2. Knowledge Acquisition System

Helps in collecting and organizing knowledge from human experts.

Provides tools and methods to properly store and structure this knowledge.
3. Inference Engine (Reasoning System)

The core component that processes information from the knowledge base.
Uses logical rules to make conclusions, provide solutions, and give recommendations.
4. User Interface

Allows users to interact with the system.

Can be a text-based interface, graphical interface (GUI), or chatbot-like system.
 5. Explanation Facility

Explains why the system reached a certain conclusion.

Makes the system more trustworthy and transparent for users.
6. Maintenance and Adaptation Tools

Helps in updating and modifying the system over time.

Allows knowledge engineers to add new information or change existing rules.

Case Studies: MYCIN and DENDRAL Expert Systems

Expert systems have been widely used in real-world applications, and two of the most famous
early expert systems are MYCIN and DENDRAL. Both were developed in the field of artificial
intelligence (AI) and demonstrated how expert systems can solve complex problems using
stored knowledge and logical reasoning.

1. MYCIN (Medical Diagnosis Expert System)

Overview:
MYCIN was an expert system developed in the 1970s at Stanford University to assist doctors
in diagnosing and treating bacterial infections, particularly blood infections (bacteremia)
and meningitis.

How It Worked:
MYCIN used a rule-based system with around 600 IF-THEN rules.
It collected information about a patient’s symptoms, lab test results, and medical history.
The inference engine applied rules to match symptoms with known diseases.
MYCIN provided diagnosis recommendations and suggested the best antibiotic treatment,
along with dosage instructions.

Example of a Rule in MYCIN:

"IF the patient has a fever AND the bacteria is gram-negative, THEN consider the possibility of
bacterial infection."

Advantages of MYCIN:
✔ More accurate than many human doctors in diagnosing infections.
✔ Provided reasoning for its recommendations, helping doctors understand why a decision
was made.
✔ Used certainty factors to handle uncertainty in medical diagnosis.

Limitations of MYCIN:
❌ Never used in real hospitals due to legal and ethical concerns (AI making medical
decisions).
❌ Could not explain its reasoning in a way that humans could always understand.
❌ Required constant updates to keep medical knowledge up to date.

2. DENDRAL (Chemical Analysis Expert System)

Overview:
DENDRAL was one of the first expert systems, developed in the 1960s at Stanford
University. It was designed to help chemists identify molecular structures of unknown
chemical compounds.

How It Worked:
Scientists provided mass spectrometry data (a method used to analyze chemical substances).
DENDRAL applied heuristic rules (expert knowledge of chemistry) to generate possible
molecular structures.
It analyzed all possible structures and selected the most likely one based on chemical rules.

Example of a Rule in DENDRAL:

"IF a compound has a mass of 180 and contains oxygen, THEN it could be a sugar molecule."

Advantages of DENDRAL:
✔ Accurately identified unknown chemical compounds faster than human experts.
✔ Helped chemists discover new molecules in scientific research.
✔ Was one of the first AI systems that showed machines could outperform humans in
specialized tasks.

Limitations of DENDRAL:
❌ Limited to organic chemistry and could not be used for other scientific domains.
❌ Required expert chemists to manually input rules, making it difficult to scale.

Both MYCIN and DENDRAL were groundbreaking expert systems that proved AI could solve
specialized problems. MYCIN showed how expert systems could assist in medical diagnosis,
while DENDRAL demonstrated AI’s power in scientific discovery. These systems paved the
way for modern AI applications in healthcare, chemistry, and beyond.

Next Topic #

← prev next →