Module 5

Chapter II- Expert system

Expert system

Expert systems are specialized systems designed to solve problems within a

narrow domain using a knowledge base and inference engine.
These systems leverage a knowledge base containing facts, rules, and
relationships, along with an inference engine to apply logical reasoning and draw
conclusions
Key Components:
Knowledge Base:
This is a structured collection of facts, rules, and relationships
related to the specific domain of expertise
Inference Engine:
This component uses the knowledge base to reason, make inferences, and
draw conclusions based on the given input.

User Interface:
This allows users to interact with the expert system, input information, and receive
outputs

Knowledge Acquisition:
This is the process of incorporating new knowledge into the knowledge base, often
through human experts,

Functionality:

Problem Solving,Decision Making,Advice Giving:

Examples:
Medical diagnosis systems,Financial investment systems,Fault diagnosis systems,Drug discovery
systems

Components and Architecture of Expert System

● Knowledge Base: The Heart of the System
● Inference Engine: The Brain Behind the Decisions
● User Interface: Bridging the Gap Between System and User
● Explanation Module: Building Trust Through Transparency
● Knowledge Acquisition Module: Keeping the System Up-to-Date.
Representing and Using Domain Knowledge
The inference engine is the brain of the expert system. It processes the information stored in
the knowledge base to draw conclusions or make recommendations. The inference engine uses
reasoning strategies (like forward chaining or backward chaining) to analyze data and apply
rules.
● Forward Chaining: Starts with available data and works toward a conclusion. For
example, "If the temperature is high and the patient has a cough, diagnose a respiratory
infection."
● Backward Chaining: Starts with a goal and works backward to find supporting
evidence. For example, "If the goal is to diagnose diabetes, check for symptoms like
frequent urination and high blood sugar."
Forward Chaining

This is a data-driven reasoning approach where the system starts with the
available facts and applies rules to infer new facts or conclusions. It's typically
used to predict outcomes or determine what will happen next.

Backward Chaining

This is a goal-driven reasoning approach where the system starts with a

hypothesis or a goal (something to prove) and works backward to determine
which facts or conditions would support that conclusion. It's often used to
diagnose issues by determining the cause of an observed effect
Types of Expert Systems in AI

. 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.

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
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.

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.
Examples of Expert Systems in AI

here have been several significant real-world expert systems developed over the years. Some of them are
given below:
1. MYCIN : As mentioned earlier, revolutionized medical diagnosis by using rule-based logic to detect
bacterial infections.
● MYCIN uses backward chaining to diagnose bacterial infections, such as meningitis and bacteremia.
It identifies the bacteria causing the infection by asking the doctor a series of questions about the
patient's symptoms and test results.
● Significance: Although not used clinically, MYCIN greatly influenced the development of medical
expert systems.
Rule 52:

If

1) THE SITE OF THE CULTURE IS BLOOD

2) THE GRAM OF THE ORGANISM IS NEG

3) THE MORPHOLOGY OF THE ORGANISM IS ROD

4) THE BURN OF THE PATIENT IS SERIOUS

Then there is weakly suggestive evidence (0.4) that

1) THE IDENTITY OF THE ORGANISM IS PSEUDOMONAS

2.PROSPECTOR is an early expert system developed by SRI International to aid
geologists in mineral exploration. It uses an inference network to evaluate the favorability
of a site for specific ore deposits. PROSPECTOR was designed to assist geologists in
decision-making during exploration, including identifying promising drilling locations

Key Features and Functionality:

Inference Network:
PROSPECTOR uses an inference network to represent the knowledge base and guide
its reasoning process.

Decision Support:
It helps geologists evaluate the potential of a site for specific types of ore deposits.
Drilling Site Selection:
PROSPECTOR can assist in identifying optimal drilling locations once a
favorable site has been identified.

Knowledge Base:
It's built upon a knowledge base of geological information, including rules and
facts about ore deposits and their geological contexts.

Rule-based Reasoning:
PROSPECTOR employs a rule-based approach to draw inferences and
evaluate the favorability of a site.
Expert system shell
An expert system shell is a pre-built framework for developing expert systems, providing the structure and
tools needed to build a system without having to start from scratch. It's essentially an empty expert system
that users can populate with their own knowledge and rules. Expert system shells offer features like
knowledge representation schemes, inference engines, user interfaces, and explanation capabilities,
making it easier to develop and deploy expert systems.

Key Characteristics of Expert System Shells:

Empty Knowledge Base:

Shells provide a framework but lack the specific knowledge of a particular domain.

Pre-built Components:
They include essential components like inference engines, user interfaces, and explanation modules.
 User-Friendly Interface:
Shells often offer a user-friendly interface for knowledge engineers to build and maintain their systems.
Knowledge Representation:
They support various knowledge representation schemes, such as rule-based, object-oriented, or
frame-based approaches.
Inference Engine:
The inference engine allows the system to reason and draw conclusions based on the knowledge base.

Explanation Module:
This module provides the ability to explain the system's reasoning process and justify its conclusions

Benefits of Using Expert System Shells

Reduced Development Time,Cost-Effective Development,Improved Maintainability:

Expert system capabilities
Explain its Reasoning

Acquire new knowledge and modify old knowledge

The knowledge acquisition program TEIRESIAS was built to assist domain experts in refining the MYCIN knowledge base.
TEIRESIAS developed the concept of metalevel knowledge, i.e., knowledge by which a program can not only use its knowledge
directly, but can examine it, reason about it, and direct its use. TEIRESIAS makes clear the line of reasoning used in making a
diagnosis and aids physician experts in modifying or adding to the knowledge base. Much of this was incorporated into the
EMYCIN framework. The flexibility and understandability that TEIRESIAS introduced into the knowledge base debugging
process have been models for the design of many expert systems. (1974-1977)
Knowledge acquisition

Knowledge acquisition in expert systems is the crucial process of extracting, structuring, and
organizing knowledge from human experts to create a system that can perform complex
problem-solving tasks. This process is essential because the effectiveness of an expert system
directly depends on the quality and completeness of the knowledge it contains.

Key aspects of knowledge acquisition in expert systems:

Purpose:
To capture and represent expert knowledge in a way that a computer can understand and use
to reason and solve problems.
Process:
It involves eliciting, analyzing, and interpreting knowledge from experts, then transforming it
into a suitable machine representation.
 Importance:
A well-acquired knowledge base is the foundation for an effective expert system, allowing it to
provide valuable insights and advice.
Techniques:
Various techniques are employed to facilitate knowledge acquisition, including interviews,
questionnaires, and observation of experts in action.
Challenges:
The "knowledge acquisition bottleneck" refers to the difficulties in obtaining, structuring, and
representing knowledge from human experts, which can be time-consuming and challenging.

In summary, knowledge acquisition is a critical phase in developing expert systems, ensuring that
the system's knowledge base is accurate, comprehensive, and suitable for reasoning and
problem-solving.
MOLE (Medical Diagnosis) and SALT (propose-and-revise)

MOLE (Medical Diagnosis) and SALT (propose-and-revise) are expert system knowledge acquisition
tools. MOLE focuses on building diagnostic systems that can handle uncertainty by burying certainty
factors, while SALT helps build systems with a propose-and-revise problem-solving approach. Both tools
automate the process of extracting and structuring knowledge from experts, making it easier to build and
refine expert systems.

MOLE (Medical Diagnosis):

● Focus: Building diagnostic expert systems.

● Uncertainty Handling: MOLE addresses the difficulty of eliciting certainty factors from experts by
integrating heuristic assumptions and generalizing expert preferences. It doesn't require explicit
certainty factors for inference rules.
● Problem-Solving Method: MOLE is used in conjunction with the "cover-and-differentiate"
problem-solving method.
SALT (Propose-and-Revise):

Focus:
Building expert systems that can propose solutions and revise them based on constraints.
Problem-Solving Strategy:
SALT is built around the "propose-and-revise" strategy, where the system incrementally
constructs a design by proposing values for design parameters and revising decisions when
constraint violations are detected.
Knowledge Representation:
SALT's knowledge representation is based on the propose-and-revise problem-solving strategy.
Knowledge Acquisition:
SALT uses its knowledge of the problem-solving strategy to identify relevant domain knowledge,
guide interrogation of experts, generate expert systems with reasoning capabilities, and analyze
test case coverage
Limitations of Expert Systems
1. Knowledge Limitation:
2. Lack of Flexibility
3. Maintenance

Expert systems are a crucial aspect of AI, providing intelligent decision-making

capabilities across various domains. By emulating human expertise, they offer
valuable insights, consistent solutions, and efficiency. Despite their limitations,
expert systems continue to evolve and play a significant role in advancing AI
technologies.