Q.1 Which of the following is a common form of knowledge representation in AI?
Decision Trees
Semantic Networks
Neural Weights
Gradient Descent
Explanation - Semantic networks are widely used for representing knowledge in AI by showing relationships between concepts.
Correct answer is: Semantic Networks
Q.2 First-order logic allows the representation of?
Only facts
Facts and relationships
Only probabilities
Neural activations
Explanation - First-order logic is powerful enough to represent both objects and their relationships using quantifiers.
Correct answer is: Facts and relationships
Q.3 Which symbol is used in predicate logic to denote 'for all'?
∃
∀
→
¬
Explanation - The universal quantifier (∀) is used to denote that a statement applies to all elements in a domain.
Correct answer is: ∀
Q.4 In knowledge representation, frames are used to?
Train neural networks
Store structured knowledge
Measure probabilities
Represent data streams
Explanation - Frames are data structures that represent stereotyped situations, helping to store structured knowledge.
Correct answer is: Store structured knowledge
Q.5 Which of these is NOT a knowledge representation technique?
Production Rules
Semantic Networks
Frames
Backpropagation
Explanation - Backpropagation is an algorithm for training neural networks, not a knowledge representation method.
Correct answer is: Backpropagation
Q.6 What does the 'closed-world assumption' imply in reasoning?
Everything not known to be true is assumed false
Everything is uncertain
All knowledge is always true
Knowledge is never complete
Explanation - Closed-world assumption assumes that if something cannot be proven true, it is false.
Correct answer is: Everything not known to be true is assumed false
Q.7 Which is a drawback of using propositional logic for knowledge representation?
Lacks inference rules
Cannot express relationships between objects
Too expressive
Not computable
Explanation - Propositional logic can only represent facts but not relationships between entities.
Correct answer is: Cannot express relationships between objects
Q.8 Which type of logic allows quantification over variables?
Propositional Logic
First-Order Logic
Boolean Algebra
Fuzzy Logic
Explanation - First-order logic introduces quantifiers like ∀ and ∃ to represent generalized statements.
Correct answer is: First-Order Logic
Q.9 Ontologies in AI are used to?
Classify data in databases
Provide structured vocabularies for domains
Train machine learning models
Store sensory data
Explanation - Ontologies define concepts, categories, and relationships within a domain, enabling structured knowledge.
Correct answer is: Provide structured vocabularies for domains
Q.10 Which representation uses 'if-then' rules?
Frames
Production Rules
Semantic Nets
Graphs
Explanation - Production rules are expressed in 'if-then' format to represent procedural knowledge.
Correct answer is: Production Rules
Q.11 Default reasoning assumes?
Facts are always true
Some facts hold unless contradicted
No facts can be assumed
Only probabilities matter
Explanation - Default reasoning allows assumptions until evidence proves otherwise.
Correct answer is: Some facts hold unless contradicted
Q.12 A truth table is commonly used in?
Frames
Propositional Logic
Semantic Networks
Neural Nets
Explanation - Truth tables are used to evaluate propositional logic expressions under different assignments.
Correct answer is: Propositional Logic
Q.13 Fuzzy logic is mainly used to handle?
Binary truth
Uncertainty and vagueness
Strict rules
Only numerical data
Explanation - Fuzzy logic represents truth values on a continuum between 0 and 1, handling vagueness.
Correct answer is: Uncertainty and vagueness
Q.14 In semantic networks, nodes represent?
Rules
Concepts
Probabilities
Algorithms
Explanation - Nodes represent concepts or objects, while edges represent relationships between them.
Correct answer is: Concepts
Q.15 Which is a limitation of semantic networks?
They can’t represent hierarchies
They can be inefficient for complex queries
They lack visual representation
They can’t handle uncertainty
Explanation - Semantic networks can grow large and become inefficient when processing queries.
Correct answer is: They can be inefficient for complex queries
Q.16 Non-monotonic reasoning allows?
Only one conclusion
Conclusions to be withdrawn when new facts arise
Infinite loops
Strict consistency
Explanation - Non-monotonic reasoning supports retracting conclusions when new evidence contradicts them.
Correct answer is: Conclusions to be withdrawn when new facts arise
Q.17 Which logic allows statements like 'Some cats are black'?
Propositional Logic
First-Order Logic
Default Logic
Fuzzy Logic
Explanation - First-order logic allows quantification to represent statements about some or all objects.
Correct answer is: First-Order Logic
Q.18 Which of the following is an example of declarative knowledge?
How to ride a bike
Paris is the capital of France
Algorithm for sorting
Steps for cooking rice
Explanation - Declarative knowledge states facts about the world, unlike procedural knowledge (how-to).
Correct answer is: Paris is the capital of France
Q.19 Which is NOT a feature of frames in knowledge representation?
Slots
Default values
Procedural attachments
Backpropagation layers
Explanation - Frames use slots, values, and procedures, while backpropagation is used in neural networks.
Correct answer is: Backpropagation layers
Q.20 The statement '∀x (Bird(x) → Fly(x))' means?
Some birds can fly
All birds can fly
No birds can fly
Birds never fly
Explanation - The universal quantifier ∀ states that for all objects x, if x is a bird, then x can fly.
Correct answer is: All birds can fly
Q.21 Reasoning based on experience is called?
Deductive reasoning
Inductive reasoning
Abductive reasoning
Default reasoning
Explanation - Inductive reasoning generalizes rules or facts from observed examples.
Correct answer is: Inductive reasoning
Q.22 Abductive reasoning is used to?
Prove theorems
Find the best explanation for observations
Deduce absolute truths
Eliminate uncertainty
Explanation - Abduction is reasoning to the best possible explanation of evidence.
Correct answer is: Find the best explanation for observations
Q.23 Which representation method is closest to human cognitive structures?
Neural Nets
Frames
Semantic Networks
Decision Trees
Explanation - Frames resemble human mental models of stereotypical situations, making them cognitively close.
Correct answer is: Frames
Q.24 Which of these is a limitation of first-order logic?
It cannot express relationships
It struggles with uncertain knowledge
It lacks inference rules
It is not formal
Explanation - First-order logic is deterministic and cannot represent uncertain or probabilistic information.
Correct answer is: It struggles with uncertain knowledge
Q.25 Probabilistic reasoning in AI deals with?
Exact truths
Uncertainty and likelihood of events
Pure logic
Neural activations
Explanation - Probabilistic reasoning estimates how likely events are when certainty is not possible.
Correct answer is: Uncertainty and likelihood of events