Q1.
What are the advantages and disadvantages of Expert System in
Artificial Intelligence? Explain in detail.
Expert systems are computer programs designed to provide expert-level knowledge
and analytical skills to emulate the decision making ability of a human expert
Some of the main advantages of expert systems include:
1. Consistency - Expert systems always respond the same way under the same
conditions, unlike human experts who may provide slightly different answers based
on factors like tiredness or emotional state. This consistency can improve reliability.
2. Availability - Expert systems are always available when needed and can handle
high volumes of repetitive decisions without fatigue. Having them available at any
time improves access.
3. Reduces costs - In many cases, expert systems reduce the need for a human expert
to be physically present to provide advice or make diagnoses. This reduces labor
costs.
4. Permanence - The knowledge contained in an expert system can persist indefinitely,
unlike human experts who retire or leave an organization. This retains valuable
expertise.
5. Explainability - Many expert systems can explain the logic and reasoning behind
their decisions and recommendations. This increases transparency.
However, expert systems also come with some disadvantages,
including:
1. Lack of common sense - Human experts have a lifetime of real-world experience
and common sense to apply to decision making, which expert systems lack. This can
limit their capability to assess some situations.
2. Brittleness - Expert systems are limited by the finite knowledge they were
programmed with, and may fail unpredictably when faced with situations outside their
domain. Human experts are generally more adaptable.
3. Development costs - A great deal of human specialist time is required for
knowledge engineering to transfer domain knowledge into a form the expert system
can apply. Development is labor intensive.
4. Maintenance challenges - The knowledge inside an expert system can become
outdated over time. But updating the knowledge base can be expensive, tricky and
time consuming. Keeping systems current is difficult.
5. Lack of human touch - The advice of a computer program may be less convincing
to some users compared to human expert counsel, empathy and non-verbal
communication. Some nuance and connection may be lost.
In summary, while expert systems offer benefits like speed, consistency and cost
savings, they struggle to match human flexibility, adaptability and the human touch.
Balancing the advantages and limitations is important for successful applications.
Q2. What is knowledge based systems? What can knowledge based
systems do? Give your answer in detail.
A knowledge based system (KBS) is a form of artificial intelligence application that
uses a knowledge base along with inference rules to provide expertise and analytical
decisions in a narrow domain. A knowledge base is a specialized database containing
the accumulated experience and domain knowledge of human experts, represented in
a machine-readable format.
Some of the major capabilities of knowledge based systems include:
1. Make complex decisions - By encoding domain expertise and rules, a KBS can
leverage facts and heuristics to make decisions that would normally require human
analysis and specialized judgment. This enables systems to make recommendations in
areas like medical diagnosis, technical troubleshooting and financial planning.
2. Respond rapidly with consistency - Unlike human decision making which is prone
to delays, distractions and emotional influences, KBSs can analyze situations and
respond immediately with the same results every time the same inputs are supplied.
This consistency improves reliability.
3. Handle risky or inaccessible tasks - Knowledge based systems enable expertise to
be applied in dangerous environments like mining operations or radioactive zones
without risking human health. They can also make knowledge available anywhere via
networks.
4. Preserve and reuse rare knowledge - The knowledge of specialists with rare
expertise can be preserved indefinitely in a KBS and reused long after those experts
are gone. This helps prevent losses of vital skills and know-how.
5. Improve training efficiency - The encoded knowledge and interactive
question/response abilities of knowledge based systems can serve as an always-
available training method to rapidly educate new personnel. This reduces demands on
expert time.
6. Explain the reason behind conclusions - Many KBSs are capable of explaining their
line of reasoning to human users, describing how they reached decisions or diagnoses
step-by-step. This increases transparency and trust in the system.
7. Never forget or overlook details - Unlike humans who can occasionally forget
details or overlook nuances, KBSs consistently take into account all encoded
knowledge and inferences relevant to each situation without fail. This strengthens
overall performance.
8. Handle very large datasets - Knowledge based systems can sift through, organize
and analyze extremely high volumes of multi-structured data that would overwhelm
human cognitive capabilities. This facilitates new insights.
In essence, knowledge based systems amplify and distribute human expertise to allow
consistent, explainable decisions and analytical results in defined domains that would
otherwise require scarce and costly specialist intervention. The combination of
encoded domain knowledge and machine reasoning offers organizations powerful
decision making and data processing capabilities.
Q3. What is Knowledge Acquisition? Define Structured Interviews in
detail:
Knowledge acquisition refers to the process of extracting, organizing, processing, and
translating expert domain knowledge into a computer-usable format that can be coded
into knowledge bases for use in knowledge-based systems and artificial intelligence
applications.
Since domain specialists often gain expertise over many years of experience and
practice rather than through formal instruction, the knowledge they have accumulated
is usually tacit and unstructured. A major challenge in knowledge engineering lies in
converting the complex experiential knowledge experts carry in their minds into an
explicit, structured representation a computer can apply reliably and consistently to
solve problems and make analytical decisions in that field.
Structured interviews serve as one of the most important knowledge
acquisition techniques for eliciting expert knowledge. In a structured
interview:
1. The knowledge engineer carefully plans a set of open-ended questions in advance
to probe different aspects of the specialist's knowledge and walk through sample cases
in the field.
2. The questions follow a defined outline to methodically cover one domain area at a
time while allowing flexibility for the specialist to fully explain concepts, illuminate
nuances, and connect insights.
3. The knowledge engineer guides interactions, tracks responses, asks follow-up
questions for clarification when needed, and summarizes conclusions to verify
understanding.
4. The interview typically lasts about two hours and may run through multiple focused
sessions over weeks to completely map the breadth and depth of the specialist's
experiential knowledge.
5. The knowledge engineer then analyzes interview transcripts to identify explicit
rules, meaningful patterns and structures within responses that can inform the
knowledge base architecture and processing mechanisms.
The goal is to direct specialists into fully verbalizing their implicit knowledge by
walking through sample cases and decision flows. This helps encode not just the
factual rules they follow but also the implicit priorities, logical chains, assumptions,
and rationales behind their expertise. The one-on-one question format can encourage
deeper explanation of domain facets missed by surveys or group techniques.
In essence, structured interviews allow knowledge engineers to capture rich
qualitative insights directly from specialists' own words and examples, for translation
into accurate, multifaceted system capabilities that can replicate human expert level
decisions. They serve as a foundational technique for unlocking and analyzing the
nuanced domain understanding residing in specialists' minds.
Refernces:
https://www.javatpoint.com/advantages-and-disadvantages-of-expert-system
https://www.techtarget.com/searchcio/definition/knowledge-based-systems-KBS
https://www.autoblocks.ai/glossary/knowledge-acquisition
