ENDSEM IMP ARTIFICIAL INTELLIGENCE UNIT – 4

Q.1] List the inference rules used in prepositional logic? Explain them in detail
with suitable example.
ANS: In propositional logic, inference rules are used to derive new propositions from
existing ones. Here's a simplified explanation of some common inference rules along
with examples:
1. Modus Ponens:
o If we have a proposition of the form "if A then B" (A implies B), and we
know that A is true, then we can infer that B is true.
o Example:
 Proposition 1: If it is raining, then the streets are wet. (A implies B)
 Proposition 2: It is raining. (A)
 Conclusion: Therefore, the streets are wet. (B)
2. Modus Tollens:
o If we have a proposition of the form "if A then B" (A implies B), and we
know that B is false, then we can infer that A is false.
o Example:
 Proposition 1: If it is raining, then the streets are wet. (A implies B)
 Proposition 2: The streets are not wet. (not B)
 Conclusion: Therefore, it is not raining. (not A)
3. Hypothetical Syllogism:
o If we have two propositions of the form "if A then B" and "if B then C",
then we can infer "if A then C".
o Example:
 Proposition 1: If it is raining, then the streets are wet. (A implies B)
 Proposition 2: If the streets are wet, then people use umbrellas. (B
implies C)
 Conclusion: Therefore, if it is raining, then people use umbrellas. (A
implies C)
4. Disjunctive Syllogism:
o If we have a disjunction (A or B), and one of the disjuncts is known to be
false (not A or not B), then we can infer the truth of the other disjunct.
o Example:
 Proposition 1: Either it is raining or the sun is shining. (A or B)
 Proposition 2: It is not raining. (not A)
 Conclusion: Therefore, the sun is shining. (B)
5. Addition:
o If we have a proposition A, we can infer the disjunction "A or B" for any
proposition B.
o Example:
 Proposition 1: It is raining. (A)
 Conclusion: Therefore, it is raining or the sun is shining. (A or B)
Q.2] Explain syntax and semantics of First Order Logic in detail.
ANS: here's a breakdown of the syntax and semantics of First Order Logic (FOL) in a
simple and easy-to-understand point-wise format:
Syntax of First Order Logic:
1. Terms:
o Terms are basic building blocks representing objects or individuals in the
domain of discourse.
o They can be constants, variables, or functions applied to arguments.
o Constants represent specific objects (e.g., "John", "Mary").
o Variables represent unspecified objects (e.g., "x", "y").
o Functions take one or more arguments and return a term (e.g.,
"FatherOf(x)").
2. Predicates:
o Predicates represent properties or relations between objects.
o They are often denoted by symbols followed by arguments within
parentheses.
o Predicates can be unary (e.g., "IsAlive(x)"), binary (e.g., "Likes(x, y)"),
or n-ary (e.g., "Owns(x, y, z)").
3. Connectives:
o Connectives are logical operators used to combine atomic formulas to
form complex formulas.
o Common connectives include "AND" (∧), "OR" (∨), "NOT" (¬), "IMPLIES"
(→), and "IF AND ONLY IF" (↔).
4. Quantifiers:
o Quantifiers are used to specify the scope of variables in formulas.
o The universal quantifier (∀) indicates that a formula holds for all objects
in the domain.
o The existential quantifier (∃) indicates that a formula holds for at least
one object in the domain.
5. Formulas:
o Formulas in FOL are constructed using terms, predicates, connectives,
and quantifiers.
o Atomic formulas consist of a predicate followed by terms.
o Complex formulas are formed by combining atomic formulas using
connectives and quantifiers.
Semantics of First Order Logic:
1. Interpretation:
o An interpretation assigns meaning to the symbols and formulas in FOL.
o It consists of a domain of discourse (set of objects) and interpretations
for predicates, functions, and constants.
2. Satisfaction:
o A formula is satisfied under an interpretation if it evaluates to true in
that interpretation.
o The satisfaction of atomic formulas depends on the interpretation of
predicates and the objects they refer to.
3. Quantifier Interpretation:
o Universal quantification (∀x) states that a formula holds for all objects
in the domain.
 o Existential quantification (∃x) states that a formula holds for at least
one object in the domain.
4. Models:
o A model is an interpretation that satisfies all the formulas in a given
logical system.
o A logical system is consistent if it has at least one model.
5. Logical Consequence:
o A formula A is a logical consequence of a set of formulas Γ if every
interpretation that satisfies all formulas in Γ also satisfies A.
o This relationship is denoted as Γ ⊨ A.
Q.3] Detail the algorithm for deciding entailment in prepositional logic.
ANS: Here's a simplified algorithm for deciding entailment in propositional logic:

1. Input Propositions: Begin with two propositions, typically labeled as P

(premise) and Q (conclusion), expressed in propositional logic. These
propositions consist of atomic statements connected by logical connectives such
as AND (∧), OR (∨), NOT (¬), IMPLIES (→), etc.

2. Construct Truth Tables: Create truth tables for both propositions P and Q. A
truth table exhaustively lists all possible truth values for the atomic
statements within each proposition, along with the truth values of compound
statements formed using logical connectives.

3. Compare Truth Values: Examine the truth values of the compound statements
in both P and Q for all possible combinations of truth values of the atomic
statements.

4. Check Entailment: Check if every row where P is true, Q is also true. If this
condition holds true for all rows where P is true, then P entails Q, denoted as
P⊨Q.

5. Determine Entailment: If the truth table indicates that P entails Q, then

conclude that P entails Q. Otherwise, if there is any row where P is true and Q
is false, then P does not entail Q.

6. Example: For example, if P=(A∧B)→C and Q=A→(B→C), construct a truth

table with columns for A, B, C, P, and Q. Then, evaluate each row to
determine if P entails Q based on the truth values.

7. Final Result: Conclude the result based on the truth table. If every row where
PP is true, Q is also true, then P entails Q. Otherwise, P does not entail Q.

8. Complexity Analysis: Note that constructing truth tables can become

impractical for complex propositions with many atomic statements, as the
number of rows in the truth table grows exponentially with the number of
atomic statements.

9. Alternative Approaches: In practice, more efficient algorithms and methods

are used to decide entailment in propositional logic, such as resolution
refutation, semantic tableaux, or automated theorem proving using
algorithms like Davis-Putnam-Logemann-Loveland (DPLL) or conflict-driven
clause learning (CDCL). These methods are more scalable and can handle
larger and more complex propositional logic problems efficiently.
Q.4] Explain knowledge representation structure and compare them.
ANS:
1. Semantic Networks:
o Represent knowledge as nodes (concepts) connected by edges
(relationships).
o Easy to understand and visualize.
o Supports inheritance (inheritance of properties from parent nodes).
o Example: Concept maps.
2. Frames:
o Organize knowledge into hierarchical structures called frames.
o Each frame consists of slots for storing properties or attributes.
o Supports default values and inheritance.
o Example: Object-oriented programming.
3. Rule-Based Systems:
o Represent knowledge as a set of rules (if-then statements).
o Infer new knowledge by applying rules to existing knowledge.
o Explicit representation of logical relationships.
o Example: Expert systems.
4. Ontologies:
o Formal representation of knowledge as a set of concepts within a
domain and the relationships between them.
o Support for defining classes, properties, and instances.
o Promotes interoperability and consistency in knowledge representation.
o Example: The Gene Ontology for molecular biology.
Comparison:
 Expressiveness: Semantic networks and frames are good for representing
hierarchical knowledge, while rule-based systems excel at representing logical
relationships. Ontologies provide a formal structure for representing complex
knowledge domains.
 Inference: Rule-based systems are strong in inference, allowing deduction of
new knowledge from existing rules. Semantic networks and frames typically
lack built-in inference mechanisms.
 Ease of Use: Semantic networks and frames are more intuitive and easy to
understand due to their visual nature. Rule-based systems require
understanding of logical rules, and ontologies can be complex to create and
maintain.
 Scalability: Ontologies are highly scalable and can represent large and
complex knowledge domains. Rule-based systems can also handle large
knowledge bases, but may suffer from performance issues with very large rule
sets.
 Interoperability: Ontologies are designed for interoperability, allowing
different systems to understand and exchange knowledge. Semantic networks
and frames may lack standardization, making interoperability more
challenging.
 Domain Specificity: Rule-based systems and ontologies are well-suited for
representing knowledge in specific domains due to their formalism and
structure. Semantic networks and frames may be more flexible but may lack
the specificity needed for certain domains.
Q.5] Explain Wumpus world environment giving its PEAS description.
ANS: here's a simple and easy-to-understand explanation of the Wumpus World
environment along with its PEAS description:
1. Percept:
o Agents in the Wumpus World receive sensory information about their
immediate surroundings.
o Percepts include information about the presence of obstacles, pits, the
Wumpus, and the smell of the Wumpus.
2. Environment:
o The Wumpus World consists of a grid-like cave system where the agent
moves around.
o The environment contains pits that the agent must avoid falling into, a
Wumpus that can eat the agent, and walls that restrict movement.
3. Actuators:
o Agents can perform actions to navigate and interact with the
environment.
o Actuators include moving forward, turning left or right, shooting arrows
to kill the Wumpus, grabbing gold, and climbing out of the cave.
4. Performance:
o The performance of the agent is evaluated based on its ability to collect
gold and return safely to the starting position without being eaten by
the Wumpus or falling into pits.
o The agent receives positive rewards for picking up gold and negative
penalties for falling into pits or being eaten by the Wumpus.

Now, let's put this in a formal PEAS description:

 Performance Measure: The agent's performance is measured by the amount of

gold it collects and its ability to return safely to the starting position without
dying.
 Environment: The environment consists of a grid-like cave system with pits, a
Wumpus, gold, and walls. The agent navigates this environment using its
actuators.
 Actuators: The agent can move forward, turn left or right, shoot arrows, grab
gold, and climb out of the cave.
 Sensors: The agent receives sensory information about its immediate
surroundings, including the presence of obstacles, pits, the Wumpus, and the
smell of the Wumpus.
Q.6] Explain different inference rules in FOL with suitable example.
ANS: here's a simple explanation of different inference rules in First-Order Logic
(FOL), along with examples:
1. Modus Ponens:
o If we have two premises, "if A then B" (A -> B) and "A is true" (A), then
we can infer that "B is true" (B).
o Example: Premises:
 If it is raining, then the streets are wet. (Raining -> WetStreets)
 It is raining. (Raining) Conclusion:
 Therefore, the streets are wet. (WetStreets)
2. Modus Tollens:
o If we have two premises, "if A then B" (A -> B) and "not B" (~B), then we
can infer that "not A" (~A).
o Example: Premises:
 If it is raining, then the streets are wet. (Raining -> WetStreets)
 The streets are not wet. (~WetStreets) Conclusion:
 Therefore, it is not raining. (~Raining)
3. Universal Instantiation:
o If we know that all members of a set have a certain property, then we
can infer that a specific member of that set also has that property.
o Example: Premise:
 All cats are mammals. (∀x Cat(x) -> Mammal(x)) Conclusion:
 Therefore, Fluffy is a mammal. (Mammal(Fluffy))
4. Existential Instantiation:
o If we know that something with a certain property exists, then we can
infer that there exists a specific instance of that property.
o Example: Premise:
 There exists a student who is good at math. (∃x Student(x) ∧
GoodAtMath(x)) Conclusion:
 Therefore, John is good at math. (GoodAtMath(John))
5. Universal Generalization:
o If we know that a specific individual has a certain property, then we can
infer that all individuals have that property.
o Example: Premise:
 Socrates is a human. (Human(Socrates)) Conclusion:
 Therefore, all humans are mortal. (∀x Human(x) -> Mortal(x))
6. Existential Generalization:
o If we know that a specific individual has a certain property, then we can
infer that there exists an individual with that property.
o Example: Premise:
 John is a doctor. (Doctor(John)) Conclusion:
 Therefore, there exists a doctor. (∃x Doctor(x))
These inference rules form the basis of reasoning in First-Order Logic and are
essential for making deductions from given premises.
Q.7] Write an propositional logic for the statement,
i) "All birds fly"
ii) "Every man respect his parents"
ANS: let's break down each statement into propositional logic:
i) "All birds fly":
1. Universal Quantifier: The statement starts with "All," which indicates that it
applies to every member of a certain group.
2. Subject: The subject of the statement is "birds," which refers to a category of
animals.
3. Predicate: The predicate is "fly," which describes an action or characteristic
attributed to the subject.
4. Universal Conditional Statement: Combining the universal quantifier with the
subject and predicate, the statement can be represented as:
o ∀x (Bird(x) → Fly(x))
 ∀x denotes "for all x," meaning this applies to every member x.
 Bird(x) represents "x is a bird."
 Fly(x) represents "x can fly."
 The arrow → indicates "implies," meaning if x is a bird, then x can
fly.
ii) "Every man respects his parents":
1. Universal Quantifier: Similar to the first statement, "Every" indicates that this
applies universally.
2. Subject: The subject is "man," referring to the category of individuals.
3. Predicate: The predicate is "respects his parents," indicating an action or
behavior.
4. Universal Conditional Statement: Representing this in propositional logic:
o ∀y (Man(y) → RespectsParents(y))
 ∀y denotes "for all y," meaning this applies to every member y.
 Man(y) represents "y is a man."
 RespectsParents(y) represents "y respects his parents."
 The arrow → indicates "implies," meaning if y is a man, then y
respects his parents.
Explanation:
 In propositional logic, universal quantifiers (∀) are used to express statements
that apply to all members of a certain group.
 The subject of the statement is represented by a variable (e.g., x, y), and
predicates are used to describe the relationship or properties of the subject.
 The arrow → represents implication, indicating that if the subject satisfies the
condition on the left side, then the predicate on the right side must also hold
true.
 These statements are universal generalizations, meaning they claim
something about every member of a specified group. In the first statement,
it's about all birds, and in the second, it's about all men.
Q.8] Differentiate between propositional logic and First order logic.
ANS: Here's a simple and easy point-wise comparison between propositional logic
and first-order logic:
1. Scope of Representation:
o Propositional Logic: Represents propositions or statements as a whole,
without breaking them down into smaller components.
o First-Order Logic: Allows for the representation of propositions in terms
of objects, relations, functions, and quantifiers, enabling a more
granular and expressive representation.
2. Atomic Units:
o Propositional Logic: Operates with atomic propositions, which cannot be
further decomposed.
o First-Order Logic: Operates with atomic formulas, which can contain
variables representing objects, enabling the representation of more
complex relationships.
3. Expressive Power:
o Propositional Logic: Limited in expressing relationships between
different objects or individuals.
o First-Order Logic: More expressive, capable of representing relationships
between objects, predicates, and quantified variables.
4. Variables and Quantifiers:
o Propositional Logic: Lacks variables and quantifiers.
o First-Order Logic: Includes variables and quantifiers like existential (∃)
and universal (∀), allowing for the formulation of statements about "all"
or "some" objects.
5. Predicates and Functions:
o Propositional Logic: Does not support predicates or functions.
o First-Order Logic: Allows the use of predicates to denote properties or
relations between objects, and functions to denote mappings between
objects.
6. Semantic Interpretation:
o Propositional Logic: Truth values (true or false) are assigned directly to
propositions.
o First-Order Logic: Interpretations involve assigning objects to variables,
truth values to atomic formulas, and satisfying quantifiers over domains
of discourse.
7. Compactness:
o Propositional Logic: Often results in large, cumbersome representations
for complex systems due to its limited expressiveness.
o First-Order Logic: Provides a more compact representation for complex
systems by allowing the use of variables and quantifiers.
8. Applications:
o Propositional Logic: Commonly used in areas like digital circuit design,
automated reasoning, and artificial intelligence.
o First-Order Logic: Widely used in various domains including
mathematics, computer science, linguistics, philosophy, and artificial
intelligence for reasoning about structured knowledge and relationships.
9. Example:
o Propositional Logic: "It is raining" (simple proposition).
o First-Order Logic: "For all x, if x is a person, then x has a mother" (using
quantifiers and predicates).
Q.9] What is an Agent. Name any 5 agents around you Explain Knowledge
based agent with Wumpus World. List and explain in short the various steps of
knowledge engineering process.
ANS: An agent is a program or system that perceives its environment through
sensors and acts upon that environment through actuators based on its observations
and goals.

Here are five examples of agents around you:

1. Thermostat: It senses the temperature of the environment and adjusts the

heating or cooling system accordingly.
2. Chatbot: It interacts with users through text or speech and provides responses
based on pre-defined knowledge or machine learning algorithms.
3. Robot Vacuum Cleaner: It navigates through a room, detects obstacles, and
cleans the floor autonomously.
4. Smartphone Assistant (e.g., Siri, Google Assistant): It processes voice
commands, searches the web, sets reminders, and performs various tasks.
5. Autonomous Car: It perceives its surroundings using sensors like cameras and
radars, processes this information to make driving decisions, and controls the
vehicle accordingly.

Now, let's explain the concept of a knowledge-based agent using the Wumpus
World example:
In the Wumpus World environment, a knowledge-based agent utilizes logical
reasoning and domain-specific knowledge to navigate and make decisions. Here's
how it works:

1. Perception: The agent perceives its environment by receiving sensory input

from the Wumpus World grid, including information about adjacent squares,
smells indicating the presence of a wumpus, and breezes indicating nearby
pits.
2. Knowledge Representation: The agent represents its knowledge about the
environment using logical statements or a knowledge base. For example, it
may represent the location of pits and the wumpus, as well as the
consequences of actions like moving forward or shooting an arrow.
3. Inference: Based on its current perceptions and existing knowledge, the agent
infers new information about the environment. It deduces the possible
locations of hazards like pits and the wumpus, as well as safe paths to explore.
4. Action Selection: Using the inferred knowledge, the agent selects actions to
perform in the environment. These actions may include moving to a
neighboring square, shooting an arrow to kill the wumpus, or grabbing the
gold if it's nearby.
5. Execution: The agent executes the selected actions by updating its state in the
environment. For example, if it moves forward, it updates its position on the
grid and receives feedback about what it perceives in the new location.
Now, let's break down the steps of the knowledge engineering process:
1. Requirement Analysis: Identify the requirements and objectives of the
knowledge-based system, including its intended functionality and the
problem domain it will address.
2. Knowledge Acquisition: Gather relevant knowledge from domain experts,
existing sources, or data repositories. This may involve interviews,
documentation review, or data extraction techniques.
3. Knowledge Representation: Choose appropriate formalisms and structures to
represent the acquired knowledge within the system. This could involve logic,
rules, ontologies, or semantic networks.
4. Knowledge Engineering: Develop algorithms, heuristics, or inference
mechanisms to manipulate and reason with the represented knowledge. This
step often involves programming and software development.
5. Knowledge Integration: Integrate the knowledge-based components into the
overall system architecture, ensuring compatibility with other modules and
interfaces.
6. Testing and Evaluation: Validate the knowledge-based system through
testing and evaluation processes. This may involve simulated environments,
real-world scenarios, or user feedback.
7. Maintenance and Update: Continuously monitor and maintain the
knowledge-based system, updating its knowledge base and algorithms as
needed to adapt to changing requirements or environments.
8. Documentation: Document the design, implementation, and usage of the
knowledge-based system to facilitate understanding, troubleshooting, and
future enhancements.
Q.10] Consider the following axioms:
If a triangle is isosceles, then its two sides AB and AC are equal,
If AB and AC are equal, then angle B and C are equal
ABC is an equilateral triangle,
Represent these facts in predicate’logic.
Explain Inference in Propositional Logic.
ANS: To represent the given facts in predicate logic:
1. Let I(x)I(x) represent "x is an isosceles triangle."
2. Let E(x)E(x) represent "x is an equilateral triangle."
3. Let S(x,y)S(x,y) represent "x and y are equal in length."
4. Let A(x,y)A(x,y) represent "x and y are adjacent angles in triangle."
Now, we can express the axioms in predicate logic:
1. If a triangle is isosceles, then its two sides AB and AC are equal:
∀x(I(x)→S(AB,AC))
2. If AB and AC are equal, then angle B and angle C are equal:
∀x,y(S(AB,AC)→A(B,C))
3. ABC is an equilateral triangle:
E(ABC)

Now, let's discuss inference in propositional logic:

Inference in propositional logic refers to the process of drawing logical conclusions
from a set of given premises using valid logical rules. Here's a simplified explanation
in point form:

1. Premises: Start with a set of premises, which are statements assumed to be

true.
2. Rules of Inference: Use valid logical rules to make deductions from the
premises.
3. Logical Connectives: Apply logical connectives such as AND (∧), OR (∨), and
NOT (¬) to manipulate statements.
4. Modus Ponens: One of the fundamental rules of inference is Modus Ponens,
which states that if we have a conditional statement of the form "if A, then B,"
and we know A is true, we can infer that B is true.
5. Modus Tollens: Another important rule is Modus Tollens, which states that if we
have a conditional statement of the form "if A, then B," and we know that B is
false, we can infer that A is false.
6. Law of Detachment: If we have a conditional statement of the form "if A, then
B," and we have A is true, we can infer B.
7. Law of Syllogism: If we have two conditional statements such as "if A, then B"
and "if B, then C," we can infer "if A, then C" by connecting the premises.
8. Example: If we have the premises "if it rains, then the ground is wet" and "it is
raining," we can infer that "the ground is wet" using Modus Ponens.
9. Validity: It's important to ensure that the rules of inference used are valid and
that the conclusions drawn logically follow from the premises.
Q.11] Write the following sentences in FOL (any 2) (using types of quantifiers).
i) Every number is either negative or has a square root .
ii) Every connected and circuit-free graph is a tree .
iii) Some people are either religious or pious
iv) There is a barber who shaves all men in the town who do not shave
themselves.
ANS: here are the sentences rewritten in First Order Logic (FOL) using different
types of quantifiers:

i) Every number is either negative or has a square root:

1. Let N(x)N(x) represent "x is a number."

2. Let Neg(x)Neg(x) represent "x is negative."
3. Let Sqrt(x)Sqrt(x) represent "x has a square root."
4. The FOL expression for the given sentence is:
∀x (N(x)→(Neg(x)∨Sqrt(x)))

ii) Every connected and circuit-free graph is a tree:

1. Let G(x)G(x) represent "x is a graph."

2. Let C(x)C(x) represent "x is connected."
3. Let F(x)F(x) represent "x is circuit-free."
4. Let T(x)T(x) represent "x is a tree."
5. The FOL expression for the given sentence is:
∀x ((G(x)∧C(x)∧F(x))→T(x))

iii) Some people are either religious or pious:

1. Let P(x)P(x) represent "x is a person."

2. Let R(x)R(x) represent "x is religious."
3. Let Pious(x)Pious(x) represent "x is pious."
4. The FOL expression for the given sentence is:
∃x (P(x)∧(R(x)∨Pious(x)))

iv) There is a barber who shaves all men in the town who do not shave themselves:

1. Let B(x)B(x) represent "x is a barber."

2. Let M(x)M(x) represent "x is a man."
3. Let S(x)S(x) represent "x shaves himself."
4. The FOL expression for the given sentence is:
∃x (B(x)∧∀y ((M(y)∧¬S(y))→S(x,y)))

These formulations represent the given statements in a concise and logical manner
using First Order Logic.
Q.12] What is Resolution? Solve the following statement by using resolution
algorithm. Draw suitable resolution graph.
i) Rajesh like all kind of food.
ii) Apple and vegetables are food.
iii) Anything anyone eats and is not killed is food.
iv) Ajay eats peanuts and still alive.
Prove that Rajesh like bananas. .
ANS: Resolution is a logical inference rule used in propositional logic and first-order
logic to prove the validity of a statement. It involves using a proof by contradiction
method to show that a statement follows logically from a set of premises.
Here's how to solve the given statement using the resolution algorithm, along with a
resolution graph:
1. Convert Statements to Clauses:
o i) Rajesh likes all kinds of food. => Rajesh likes all kinds of food.
o ii) Apple and vegetables are food. => Food(apple) ∧ Food(vegetables).
o iii) Anything anyone eats and is not killed is food. => Eats(x, y) ∧
¬Killed(y) => Food(y).
o iv) Ajay eats peanuts and is still alive. => Eats(Ajay, peanuts) ∧
¬Killed(Ajay).
2. Convert to Clausal Form:
o i) Rajesh likes all kinds of food. => ¬Likes(Rajesh, x) ∨ Food(x).
o ii) Food(apple) ∧ Food(vegetables).
o iii) Eats(x, y) ∧ ¬Killed(y) => Food(y).
o iv) Eats(Ajay, peanuts) ∧ ¬Killed(Ajay).
3. Negate the Conclusion:
o We want to prove that Rajesh likes bananas, so we negate it:
¬Likes(Rajesh, bananas).
4. Apply Resolution Algorithm:
o Combine all clauses and the negation of the conclusion.
o Look for a contradiction.
5. Resolution Steps:
o Start with the initial clauses and apply resolution to derive new clauses
until a contradiction or the conclusion is reached.
6. Resolution Graph:
o Draw a graph where each node represents a clause.
o Connect nodes with edges representing the resolution steps.
7. Conclusion:
o If the contradiction is reached, then the negation of the conclusion is
proved.
o If the conclusion is reached, then the statement is proved.
8. Result:
o After applying resolution, if we reach the clause ¬Likes(Rajesh,
bananas), then we have proved that Rajesh likes bananas. If we reach a
contradiction, then it implies that Rajesh does not like bananas.
DIAGRAM :
¬Likes(Rajesh, bananas)
/ \
Likes(Rajesh, x) ¬Food(bananas)
/ \
Food(x) Food(bananas)
/\ |
Food(apple) Food(vegetables)
|
Eats(x, y) ∧ ¬Killed(y)
|
Eats(Ajay, peanuts) ∧ ¬Killed(Ajay)
Q.13] What is an Agent. Name any 5 agents around you Explain Knowledge
based agent with Wumpus World. List and explain in short the various
steps of knowledge engineering process
Consider the following axioms: If a triangle is equilateral then it is isosceles.
ANS: Agent:
An agent is a system or entity that perceives its environment through sensors and
acts upon that environment through actuators, based on its observations and
internal state. Agents are commonly used in artificial intelligence and robotics to
accomplish tasks autonomously or semi-autonomously.
Examples of Agents:
1. Chatbots: These are agents designed to interact with users through text or
speech, often used for customer service or information retrieval.
2. Autonomous Vehicles: Self-driving cars are agents that perceive their
surroundings using sensors like cameras and lidar, and make decisions on
steering, acceleration, and braking to navigate safely.
3. Virtual Assistants: Agents like Siri, Google Assistant, and Alexa respond to
voice commands and perform various tasks such as setting reminders, playing
music, or providing information.
4. Robotic Arms: Industrial robots equipped with sensors and actuators to
manipulate objects in manufacturing processes.
5. Search Algorithms: Web crawlers used by search engines to discover and index
web pages, based on predefined rules and algorithms.

Knowledge-Based Agent with Wumpus World:

In the context of the Wumpus World, a knowledge-based agent is an intelligent
agent that operates in a simulated environment where it must navigate and make
decisions based on available information and knowledge. In Wumpus World, the
agent explores a grid-based environment with pits, a wumpus (a monster), and
treasure, using logical reasoning to make decisions.

Steps of Knowledge Engineering Process:

1. Requirement Analysis: Identify the goals and objectives of the system, as well
as the knowledge sources and constraints.
2. Knowledge Acquisition: Gather knowledge from experts, documents,
databases, or other sources relevant to the problem domain.
3. Knowledge Representation: Choose appropriate formalisms to represent the
acquired knowledge, such as rules, ontologies, frames, or graphs.
4. Knowledge Engineering: Design and implement the knowledge-based system,
including encoding the acquired knowledge into a form understandable by
the system.
5. Knowledge Validation: Test and evaluate the system to ensure that it behaves
correctly and produces the desired results.
6. Knowledge Maintenance: Continuously update and refine the knowledge base
to adapt to changing requirements or new insights.
Axioms:
1. If a triangle is equilateral, then it is isosceles.
Explanation:
 An equilateral triangle is a triangle in which all three sides are of equal
length.
 An isosceles triangle is a triangle with at least two sides of equal length.
 The axiom states that any equilateral triangle is also isosceles, which is true
because in an equilateral triangle, all sides are equal, meeting the definition
of an isosceles triangle. Therefore, this axiom is true in Euclidean geometry.
Q.14] Write the following sentences in FOL(using types of quantifiers)
i) All birds fly
ii) Some boys play cricket
iii) A first cousin is a child of a parent’s sibling
iv) You can fool all the people some of the time and some of the
people all the time, but you cannot fool all the people all the time
ANS: let's represent each of these sentences in First-Order Logic (FOL) using
quantifiers and explain them in detail:
i) All birds fly:
 Predicate: F(x) represents "x flies".
 Domain: Let xx be any object, and B(x) represent "x is a bird".
 FOL Representation: ∀x(B(x)→F(x))
 Explanation:
1. ∀x: This quantifier states "for all x" or "for every x", where x represents
any object.
2. B(x)→F(x): This is the implication that if xx is a bird, then x flies. It
means every object that falls under the domain of birds flies.
ii) Some boys play cricket:
 Predicate: P(x) represents "x plays cricket".
 Domain: Let xx be any object, and B(x) represent "x is a boy".
 FOL Representation: ∃x(B(x)∧P(x))
 Explanation:
1. ∃x: This quantifier states "there exists an x" or "some x", meaning at least
one object exists that satisfies the condition.
2. B(x)∧P(x): This conjunction states that the object xx is both a boy and
plays cricket. Thus, there exists at least one boy who plays cricket.
iii) A first cousin is a child of a parent’s sibling:
 Predicate: C(x,y) represents "x is a first cousin of y".
 Domain: Let x and y be any objects representing individuals.
 FOL Representation: ∀x∀y(C(x,y)↔(∃z(P(z,x)∧P(z,y)∧x≠y)))
 Explanation:
1. ∀x∀y: This quantifier states "for all x and for all y".
2. C(x,y)↔(∃z(P(z,x)∧P(z,y)∧x≠y)): This bi-conditional statement states
that x is a first cousin of y if and only if there exists a third person, z, who
is a sibling of both x and y, and x and y are not the same person.
iv) You can fool all the people some of the time and some of the people all the time,
but you cannot fool all the people all the time:
 Predicate: F(x,y) represents "x can fool y".
 Domain: Let xx and yy be any objects representing individuals.
 FOL Representation: ∃x∀y(F(x,y)∧(∃t(T(t)∧P(y,t)))) where T(t) represents "t is
some of the time".
 Explanation:
1. ∃x: This quantifier states "there exists an x" or "some x", implying that
there is at least one person who can fool others.
2. ∀y: This quantifier states "for all y", meaning that the statement applies
to all individuals.
3. F(x,y)∧(∃t(T(t)∧P(y,t))): This conjunction states that person xx can fool
person yy and there exists some time tt where yy can be fooled by xx.
This captures the idea that all people can be fooled some of the time.
Q.15] What is Knowledge Representation using propositional Logic? Compare
propositional and predicate Logic.
ANS: here's a simple and easy point-wise comparison of propositional and predicate
logic, focusing on knowledge representation:
Propositional Logic:
1. Basic Units: Propositional logic deals with propositions, which are statements
that are either true or false.
2. Syntax: It uses logical connectives such as AND (∧), OR (∨), NOT (¬),
IMPLICATION (→), and BICONDITIONAL (↔) to combine propositions.
3. Knowledge Representation: Propositional logic is limited in its ability to
represent complex relationships and structures. It's suitable for representing
simple facts or relationships where the truth value of each proposition is clear
and fixed.
4. Example:
o Proposition: "It is raining."
o Compound Proposition: "It is raining and the sun is shining."
Predicate Logic:
1. Basic Units: Predicate logic deals with predicates, which are statements that
can have variables and quantifiers.
2. Syntax: It includes variables, constants, predicates, quantifiers (such as ∀ for
universal quantification and ∃ for existential quantification), and logical
connectives similar to propositional logic.
3. Knowledge Representation: Predicate logic allows for more complex
representation of knowledge by introducing variables and quantifiers,
enabling the representation of relationships between objects and properties.
4. Example:
o Predicate: "x is a student."
o Compound Predicate: "For all x, if x is a student, then x attends
lectures."
Comparison:
1. Expressiveness: Predicate logic is more expressive than propositional logic
because it allows for the representation of relationships and properties
involving variables and quantifiers.
2. Complexity: Propositional logic is simpler and more straightforward compared
to predicate logic, which can handle more complex relationships but requires
understanding of quantifiers and variables.
3. Scope: Propositional logic is suitable for simple, fixed truths, while predicate
logic is better suited for representing more nuanced and variable
relationships.
4. Applications: Propositional logic is commonly used in simple decision-making
systems or where the knowledge domain is limited and well-defined.
Predicate logic finds applications in areas such as artificial intelligence,
natural language processing, and database querying, where complex
relationships need to be represented and reasoned about.