Languages

PROLOG

PROLOG (Programming in Logic) is a declarative logic programming language. Instead of telling the computer *how* to solve a problem, you describe *what* the problem is and *what relationships* exist, and PROLOG's inference engine attempts to find solutions.

Declarative Paradigm: Programs consist of facts and rules.
- Facts: Statements about objects and their relationships (e.g., parent(john, mary).).
- Rules: General statements about relationships (e.g., grandparent(X, Z) :- parent(X, Y), parent(Y, Z).).
Unification: The core mechanism where PROLOG matches patterns by assigning values to variables.
Backward Chaining (Goal-Driven): PROLOG's inference engine works by trying to prove a query (goal). It starts with the query and attempts to find rules whose heads match, then recursively tries to satisfy subgoals. Uses backtracking if a path fails.
Key Features: Built-in Backtracking, Pattern Matching, Recursion.
Applications: Expert systems, natural language processing, theorem proving.

% Facts
father(john, mary).
father(john, tom).
mother(susan, mary).
mother(susan, tom).

% Rules
parent(X, Y) :- father(X, Y).
parent(X, Y) :- mother(X, Y).
grandparent(X, Z) :- parent(X, Y), parent(Y, Z).

% Query
?- grandparent(X, tom).
% Expected Output: X = john ; X = susan.

Conceptual Graphs

Conceptual Graphs (CGs) are a knowledge representation formalism based on Peirce's existential graphs and semantic networks. They provide a graphically oriented, logic-based notation for representing knowledge that aims to be both human-readable and computationally tractable.

Components: A CG is a finite, connected, bipartite graph consisting of two types of nodes:
- Concept Nodes: Represent entities, attributes, states, or events. Drawn as rectangles (e.g., [Cat], [Person: John]).
- Relation Nodes: Represent relationships between concepts. Drawn as ovals or circles (e.g., (agent), (obj)).
Basic Structure: Concepts are linked to relations, and relations are linked to concepts.
- Example: "A cat is on a mat." $\rightarrow$ [Cat]-(on)-[Mat]
- Example: "John is eating an apple." $\rightarrow$ [Person: John]-(agent)-[Eat]-(object)-[Apple]
Referent Field: Concept nodes can specify particular individuals (e.g., [Person: John]) or quantities.
Operations: CGs support formal operations for reasoning: Projection, Join, Simplification.
Advantages: Expressiveness, Visual Clarity, Formal Basis.
Applications: Knowledge management, natural language understanding, semantic web.