LOGIC IN AI

LOGIC AND REPRESENTATION

PREPOSITIONAL LOGIC / PROPOSITIONAL LOGIC
PREDICATE LOGIC
OTHER LOGICS
FIRST ORDER LOGIC
SEMANTIC NETS
OTHER KNOWLEDGE REPRESENTATION SCHEMES
 LOGIC
AND
REPRESENTATION
LOGIC & REPRESENTATION
Logic and representation are fundamental concepts in artificial intelligence (AI) that play a crucial role in
modeling and solving complex problems.
LOGIC
Logic is the foundation of AI reasoning. It provides a formal framework for representing and reasoning
about knowledge.
❖ Prepositional Logic
❖ Predicate Logic
REPRESENTATION
Representation in AI refers to how data and knowledge are structured and encoded for use by artificial
intelligence systems. It involves selecting a suitable format or model to represent information, which can
be textual, numerical, symbolic, or even sensory data like images and audio. Effective representation is
crucial because it determines an AI system's ability to understand, reason, and make decisions.
PROPOSITIONAL LOGIC
Propositional logic deals with propositions or statements that can be either true or false. It uses logical
operators such as AND, OR, NOT, and implications (IF-THEN) to express relationships between
propositions. Propositional logic is often used in AI for simple decision-making and problem-solving.
PREDICATE LOGIC
Predicate logic, also known as First Order Logic (FOL) or First Order Predicate Logic, is a formal system
used in mathematics, philosophy, and artificial intelligence to express and reason about relationships
between objects in a more detailed and expressive way than propositional logic. It extends propositional
logic by introducing variables, quantifiers, predicates, and functions. Predicate logic is a powerful and
widely used tool for knowledge representation, formalizing reasoning, and expressing complex
relationships.
PROPOSITIONAL
LOGIC
PROPOSITIONAL LOGIC
One of the simplest forms of logic representation is propositional logic.
Proposition mean sentence or a statement
Logic means argument or reasoning
Propositional logic will be either True or False not both
PREDICATE
LOGIC
PREDICATE LOGIC
Predicate logic, also known as first-order logic (FOL), is a formal system used in artificial intelligence (AI)
and various other fields to represent and reason about knowledge in a more expressive and structured
manner than propositional logic. Predicate logic extends the capabilities of propositional logic by
introducing variables, quantifiers, predicates, and functions, making it suitable for representing more
complex relationships and domain-specific information.
Here are some key aspects of predicate logic in the context of AI:
Variables: Predicate logic includes variables, which represent objects in a domain without specifying their
specific values. Variables are denoted by symbols like "x," "y," and "z.“
Predicates: Predicates represent properties, relationships, or conditions that can be true or false for objects
in the domain. Predicates are often represented by uppercase letters. For example, "P(x)" could represent
the predicate "x is a prime number."
Quantifiers: Predicate logic uses quantifiers to specify the scope of variables in statements. The two most
common quantifiers are:
➢ Universal Quantifier (∀): Represents "for all" or "for every." For example, "∀x P(x)" means "for all
x, P(x) is true.“
➢ Existential Quantifier (∃): Represents "there exists." For example, "∃x P(x)" means "there exists at
least one x for which P(x) is true."
Functions: Predicate logic includes functions, which map objects or variables to other objects. Functions
are used to express relationships and operations. For instance, "f(x) = 2x" represents a function that
doubles the value of x.
Statements and Formulas: In predicate logic, statements or formulas are constructed using combinations
of variables, predicates, quantifiers, and logical connectors (AND, OR, NOT). These formulas allow you to
express complex relationships and rules.
Example in Predicate Logic: Let's consider an example of predicate logic in AI
Predicates:
P(x) represents "x is a student," Q(x) represents "x is a teacher," and R(x, y) represents "x is taught by y."
Universal Quantifier:
∀x ∀y (P(x) → R(x, y)) means "For all x and y, if x is a student, then x is taught by y."
In this example, we use variables, predicates, quantifiers, and implication to represent a rule that specifies
that every student is taught by someone.

Predicate logic is widely used in AI for knowledge representation, rule based systems, expert systems,
natural language processing, and reasoning tasks. It allows for the formalization and manipulation of
structured knowledge, making it a fundamental component of many AI applications.
OTHER
LOGIC
OTHER LOGIC
Modal Logic: Modal logic extends classical logic to deal with modalities such as necessity and possibility.
It is used in AI for reasoning about knowledge, belief, and time.
Description Logic: Description logic is a family of logics used in knowledge representation and
reasoning, often for constructing ontologies in the Semantic Web, as seen in the Web Ontology Language
(OWL).
Fuzzy Logic: Fuzzy logic deals with uncertainty and approximate reasoning. It allows for degrees of truth
between true and false and is used in AI for tasks where imprecision and vagueness are prevalent, such as
in control systems and decision support.
Default Logic: Default logic is a non-monotonic logic that allows for the definition of default rules and
exceptions, making it suitable for reasoning with incomplete information.
Temporal Logic: Temporal logic is used to reason about time and temporal relationships. It's crucial for AI
applications involving scheduling, planning, and reasoning about events over time.
Default Logic: Default logic is a non-monotonic logic that allows for the definition of default rules and
exceptions, making it suitable for reasoning with incomplete information.
Epistemic Logic: Epistemic logic is used to model knowledge and belief, especially in multi agent
systems, where agents have limited and potentially different knowledge about a shared environment.
Paraconsistent Logic: Paraconsistent logic is designed to handle contradictions and inconsistencies in
knowledge bases without leading to trivial conclusions. It's used in AI to reason in the presence of
conflicting information.
Belief Revision and Update Logic: These logics are used to model changes in beliefs and knowledge
states over time as new information is acquired or beliefs evolve.
Action Logic: Action logic is used in AI planning systems to represent actions, their preconditions, effects,
and temporal relationships, allowing for automated planning and scheduling.
Dynamic Logic: Dynamic logic is used for reasoning about changes over time, particularly in systems
involving dynamic behavior or state transitions.
Probabilistic Logic: Probabilistic logic combines logic and probability theory to represent and reason
about uncertain information. It includes Bayesian networks, Markov logic networks, and probabilistic
graphical models.
Intuitionistic Logic: Intuitionistic logic is a constructive logic that differs from classical logic by rejecting
the principle of the excluded middle and is used in some AI systems for formal verification and theorem
proving.
FIRST ORDER
LOGIC
FIRST ORDER LOGIC
First Order Logic (FOL), also known as predicate logic, is a crucial and widely used formalism in artificial
intelligence (AI) and knowledge representation. FOL extends propositional logic by introducing variables,
quantifiers, predicates, and functions, allowing for more expressive and structured representations of
knowledge and reasoning. First-order logic provides a structured and formal framework for representing
knowledge and reasoning in AI.
It is used for various AI tasks and applications, including:
Knowledge Representation: FOL is used to represent structured knowledge about the world, including
facts, relationships, and rules. This knowledge can be used by AI systems for reasoning and decision-
making.
Inference and Reasoning: FOL allows AI systems to perform logical inference, answer queries, and draw
conclusions based on the knowledge represented in FOL.
Expert Systems: Expert systems use FOL to encode expert knowledge and perform rule-based reasoning,
making them valuable for applications like medical diagnosis and troubleshooting.
Expert Systems: Expert systems use FOL to encode expert knowledge and perform rule-based reasoning,
making them valuable for applications like medical diagnosis and troubleshooting.
Natural Language Processing: FOL is used to semantically parse and represent the meaning of natural
language sentences, enabling AI systems to understand and reason about text.
Semantic Web: FOL based languages, such as OWL (Web Ontology Language), are used to model
ontologies and represent structured data on the Semantic Web.

First Order Logic expressiveness and ability to capture complex relationships make it a powerful tool for
knowledge representation and reasoning in AI systems.
SEMANTIC
NET
SEMANTIC NETWORK
Semantic networks, also known as semantic nets, are a knowledge representation technique used in
artificial intelligence (AI) and cognitive science to organize and represent knowledge in a structured and
easily understandable manner. Semantic networks are based on the concept of nodes and links, which
represent entities or concepts and their relationships. They are used to model and reason about information,
facts, and knowledge in a way that resembles a network or graph.
Key features of semantic networks in AI include
Nodes: Nodes represent entities, objects, or concepts in the domain being modeled. Each node typically
corresponds to a unique concept or piece of knowledge. Nodes can be labeled with meaningful names or
symbols.
Links: Links, also known as arcs or edges, connect nodes to represent relationships or associations
between concepts. Links have labels that describe the nature of the relationship, such as "is-a," "part-of,"
"causes," or any other relevant link type.
Hierarchical Structure: Semantic networks often exhibit a hierarchical structure, with nodes organized in
a tree-like fashion to represent hierarchical relationships. For example, in a semantic network for animals,
"mammals" might be a parent node to "cats" and "dogs.“
Attributes: Nodes in semantic networks may have associated attributes or properties that provide
additional information about the concepts they represent. Attributes can be attached to nodes to describe
characteristics, features, or facts related to the entity.
Inheritance: One of the key features of semantic networks is inheritance, which allows properties and
relationships to be inherited by child nodes from parent nodes. This mechanism simplifies the
representation of knowledge by avoiding redundancy. For instance, if "birds" are described as having
"feathers," then "sparrows" and "eagles" inherit the attribute of having feathers.
Directed Acyclic Graphs (DAGs): Semantic networks are typically organized as directed acyclic graphs
(DAGs), meaning there are no cycles in the network structure. This ensures that relationships do not form
circular dependencies.
SEMANTIC NETWORKS ARE USED IN AI FOR VARIOUS PURPOSES
Knowledge Representation: Semantic networks provide a way to represent structured knowledge about a
domain, making it easier for AI systems to access and reason with this knowledge.
Expert Systems: Expert systems often use semantic networks to model expert knowledge and rules, which
are then used for problem-solving and decision-making.
Natural Language Processing: In natural language processing, semantic networks can be used to extract
structured information from unstructured text, aiding in information retrieval and question answering.
Semantic Web: Semantic networks are an integral part of the Semantic Web, where they serve as the
foundation for ontologies and linked data, enabling machines to understand and exchange data on the web
more effectively.
Conceptual Modeling: In database design and conceptual modeling, semantic networks can be used to
represent the structure and relationships within a database schema.
Semantic networks are a flexible and intuitive way to represent knowledge, particularly when the
relationships between concepts are essential for reasoning and decision-making in AI systems.
OTHER KNOWLEDGE
REPRESENTATION
SCHEMES
KNOWLEDGE
Knowledge is understanding and awareness of something. It refers to the information, facts, skills, and
wisdom acquired through learning and experiences in life. Knowledge is a very wide concept and has no
end. Acquiring knowledge involves cognitive processes, communication, perception, and logic.
Kinds of knowledge which needs to be represented in AI systems
Object: All the facts about objects in our world domain. E.g., Guitars contains strings etc.
Events: Events are the actions which occur in our world.
Performance: It describe behavior which involves knowledge about how to do things.
Meta knowledge: It is knowledge about what we know.
Facts: Facts are the truths about the real world and what we represent.
Knowledge Base: The central component of the knowledge-based agents is the knowledge base. It is
represented as knowledge base. The Knowledge base is a group of the Sentences (Here, sentences are used
as a technical term and not identical with the English language).
KNOWLEDGE REPRESENTATION
Humans are best at understanding, reasoning, and interpreting knowledge. Human knows things, which is
knowledge and as per their knowledge they perform various actions in the real world.
But how machines do all these things comes under knowledge representation and reasoning.
Knowledge representation is a fundamental concept in artificial intelligence (AI) that
involves creating models and structures to represent information and knowledge in a way that intelligent
systems can use.
Knowledge representation and Knowledge representation reasoning is the part of Artificial
intelligence which concerned with AI agents thinking and how thinking contributes to intelligent behavior
of agents. t is also a way which describes how we can represent knowledge in artificial intelligence.
Knowledge representation is not just storing data into some database, but it also enables an
intelligent machine to learn from that knowledge and experiences so that it can behave intelligently like a
human.
OTHER KNOWLEDGE REPRESENTATION SCHEMES
1) Symbolic Representation
2) Semantic Networks
3) Frames and Objects
4) Ontologies
5) Rule Based Systems
6) Knowledge Graphs
7) Bayesian Networks and Probabilistic Models
8) Neural Networks (NN)
9) Conceptual Graphs
10) Textual Representation
11) Image and Audio Representations
THANK YOU