Font size
WorksheetsAI Systems and Applications Worksheet
Total questions: 100
Worksheet time: 53mins
A supermarket uses a system to recommend products based on your past purchases. This system is:
Non-AI
AI using pattern recognition
A basic calculator
Manual
A smart thermostat adjusts the room temperature based on your daily routine. This demonstrates:
Fixed automation
AI learning behavior from data
Manual control only
Random decision
A bank uses AI to detect fraudulent transactions by analyzing spending patterns. This is an example of:
Data entry automation
AI for anomaly detection
A digital calculator
Simple FSM
A GPS navigation app updates your route in real time based on traffic. This system is:
Non-AI
AI with adaptive behavior
Manual
Static programming
A coffee machine that starts brewing at the same time daily is:
AI
Non-AI automated system
Learning algorithm
Reinforcement learning system
A healthcare app predicts your risk of diabetes using your medical records. This AI application relies on:
Automation
Machine learning
FSM
Random guessing
An AI-powered chatbot on a customer service website improves its responses over time. This illustrates:
Non-AI automation
AI learning from interactions
Fixed rules
Data storage
AI-generated music can create new songs by learning styles of famous artists. The ethical concern here is:
Memory usage
Copyright and ownership
Audio quality
File size
A smart refrigerator alerts you when items are low based on usage patterns. This system demonstrates:
AI and pattern recognition
Random alerts
Manual checking only
Simple timers
A wearable fitness tracker adjusts workout recommendations based on your activity history. This is:
AI
Random alerts
Simple timing checks only
Manual system
An online learning platform recommends courses based on what other students in your field liked. This AI technique is:
Sorting algorithm
Collaborative filtering (pattern recognition)
Fixed automation
Data compression
An autonomous vacuum cleaner avoids obstacles and maps rooms over time. This demonstrates:
Random movement
AI with environment adaptation and learning
Timer-based cleaning
Manual control
A social media platform flags inappropriate content automatically using AI. This is:
Manual moderation
AI for pattern recognition
Fixed rule-based automation only
Hardware control
A company uses AI to predict which employees may leave the company. This requires:
Pattern recognition in employee data
Random prediction
Fixed schedules
FSM only
AI-powered translation apps improve over time by learning from user corrections. This illustrates:
Non-learning automation
Supervised machine learning
Random outputs
Hardware optimization
A delivery company uses an algorithm to optimize routes for multiple drivers daily. What is essential for this algorithm?
It must terminate and give a solution (finite)
It can run infinitely
It must be random
It ignores data
A recipe app suggests ingredient quantities based on past user ratings. This is an example of:
Algorithmic reasoning
Random suggestion
FSM only
Manual calculation
A grocery store uses “look-and-say” sequence logic to detect repeated purchase patterns. What AI concept is being used?
Recursion
Pattern recognition
Data deletion
Encryption
A warehouse robot navigates aisles using a series of programmed steps that repeat until a task is done. This illustrates:
Fixed automation
AI adapting to user data
Timer-based system
Manual logging
An online quiz app randomizes questions but ensures no repeats. Which property of an algorithm does this demonstrate?
A. Definiteness and termination
B. Infinite loop
C. Random memory
D. Automation only
A chess AI evaluates possible moves ahead to decide its next move. This is an example of:
Sequential automation
Recursive search algorithm
Random guessing
Timer-based control
A school management system automatically generates timetables each term. Which algorithmic principle ensures it completes correctly?
Finiteness
Randomization
Infinite recursion
FSM
A bank’s fraud detection system learns to identify unusual patterns in transactions. This is an example of:
Pattern recognition and machine learning
Timer-based checks
Manual auditing
FSM
A grocery delivery AI predicts future orders using past purchase sequences. Which AI concept is used?
Look-and-say sequence (pattern recognition)
BFS only
DFS only
Manual counting
Traffic lights at a busy intersection change according to sensors detecting cars. What type of AI behavior model is used?
FSM
Random automation
Pattern recognition only
Recursive search
A drone flying in a warehouse adjusts its path when obstacles appear. This demonstrates:
Adaptive AI using real-time feedback
Fixed automation
Timer-based movement
Manual control
An AI-powered email system marks spam using training data. This is:
Supervised learning
Random filtering
Fixed automation
FSM
Self-driving cars maintain safe distance using sensors and historical patterns. Which AI concept is applied?
Pattern recognition and decision-making
Timer-based checks
Fixed automation
FSM
Online exam proctoring AI detects suspicious activity like looking away frequently. This is:
Image recognition in real time
FSM only
Random alerts
Timer-based automation
A health app recommends diet changes based on your daily activity. Which AI principle is at work?
Learning from past data (pattern recognition + supervised learning)
Random advice
Fixed schedule
FSM only
A robot vacuum cleans your house and maps rooms while avoiding obstacles. Which search algorithm concept is applied?
BFS (explores level by level)
DFS (explores deep first)
Random search
Timer-based automation
GPS navigation apps use BFS to:
Find the shortest path in unweighted city roads
Explore all highways randomly
Avoid traffic
Store previous routes only
A delivery drone checks all nearby streets systematically before moving deeper into the city. Which method is this?
BFS
DFS
Random traversal
FSM
DFS might be used in maze-solving robots because:
It is memory-efficient and explores one path deeply
It guarantees shortest path always
It uses level-order exploration
It is random
A BFS algorithm is preferable in:
Urban road navigation to guarantee shortest path
Maze exploration where memory is limited
Random cleaning
Timer-based automation
A warehouse AI explores all aisle options before deeper paths. This is:
BFS
DFS
FSM
Random search
In BFS, memory usage is higher because:
It stores all frontier nodes
It stores one path only
It ignores previous nodes
It compresses data
DFS is preferable when:
Memory is limited, and depth exploration is sufficient
Shortest path is needed
Level-order traversal is required
Random search is needed
A drone exploring multiple floors of a building uses DFS for:
Deep first path exploration
Level-order traversal
Random scanning
Timer-based path
BFS guarantees the shortest path in:
Unweighted graphs
Weighted graphs
Cyclic graphs only
Directed graphs
A GPS app must frequently check all neighbors before moving further. Which algorithm does it implement?
BFS
DFS
FSM
Timer automation
DFS may fail in infinite road loops because:
It can get stuck exploring one path deeply
BFS is memory-intensive
FSM fails
Random search is slower
A robot exploring a warehouse floor level by level implements:
BFS
DFS
Random traversal
Timer automation
BFS memory requirement increases with:
Number of nodes at each level
Depth of a single path
Number of robots
Timer settings
Graph search in real-world AI is important because:
It allows autonomous navigation, logistics, and path planning
It compresses data
It encrypts signals
It reduces memory usage
A traffic light system cycles through Red → Green → Yellow repeatedly. This is modeled as:
FSM
Random automation
Recursive search
Timer-based automation
Your smart washing machine progresses through Wash → Rinse → Spin → Off. This is an example of:
FSM
Random steps
BFS
DFS
Elevator control systems that move between floors based on button input are modeled using:
FSM
AI pattern recognition
Reinforcement learning
Random automation
A turnstile that unlocks with a ticket input and locks again after passage demonstrates:
FSM
Random automation
Recursive behavior
BFS
FSMs in vending machines allow:
Sequential operations like coin → selection → dispense
Random dispensing
AI learning automatically
Memory storage
FSMs in digital devices differ from AI because:
They do not learn or adapt
They always predict
They reduce memory usage
They are slow
Home security systems that respond to sensor triggers (door open → alarm) operate as:
FSM
BFS
DFS
Random automation
FSMs are widely used because they:
Simplify control of sequential operations
Learn from data
Require complex memory
Compress signals
FSMs cannot:
Learn from experience
Control sequential events
Respond to input
Predict outputs
FSM transition occurs when:
An input/event triggers movement from one state to another
Random timing occurs
Memory is cleared
AI predicts next step
A bank AI predicts loan approval based on applicant history. The model may overfit if:
It memorizes only the training dataset without generalizing
It uses test data correctly
It simplifies features
It uses reinforcement learning
Netflix recommends movies based on your viewing history. This illustrates:
Pattern recognition and supervised learning
FSM only
Timer-based suggestions
Random selection
A school AI predicts students at risk of failing. Which ML concept is used?
Supervised learning with labeled past data
DFS
Random automation
FSM
A fitness app predicts calorie needs using past activity. Overfitting occurs if:
Predictions match training data but fail on new users
It generalizes well
Data is small
FSM is used
Cross-validation helps:
Evaluate how well an AI model will perform on unseen data
Reduce features
Encrypt outputs
Store predictions
Feature selection ensures:
Only relevant input data is used, improving performance
All inputs are always used
Memory is reduced only
Random prediction
Reducing overfitting in a real-world health AI system can be done by:
Using simpler models or more data
Memorizing training set
Ignoring test data
Increasing model complexity
In supervised learning, input variables such as age, income, and credit history are called:
Features
Labels
FSM states
Nodes
Reinforcement learning in a game app rewards the player’s AI for:
Winning strategies
Random movement
Fixed rules
BFS
Online shopping recommendation engines use:
Pattern recognition and supervised learning
FSM only
Random suggestion
Timer-based automation
Reducing false positives in medical AI involves adjusting:
A. Classification thresholds
B. Memory
C. Features only
D. FSM states
AI-driven spam filters improve over time using:
Feedback (supervised learning)
Random deletion
Timer rules
BFS
Image recognition in security cameras uses:
Feature extraction + supervised learning
FSM only
A fitness AI app learns from multiple users to suggest better workouts. This is:
Generalization through pattern recognition
FSM
Random suggestion
BFS
Overfitting in AI occurs most when:
Training data is small and too specific
Features are generalized
Test data is large
Reinforcement learning is applied
Feature engineering improves AI by:
Transforming raw data into meaningful inputs
Deleting test data
Increasing FSM states
Random prediction
Cross-validation helps in real life by:
Detecting overfitting and improving real-world accuracy
Reducing file size
Encrypting predictions
Compressing features
Email AI predicting spam uses:
Labeled past messages to learn patterns
Random deletion
FSM only
Timer rules
Stock market AI uses supervised learning to:
Predict trends based on historical patterns
Randomly choose stocks
Use FSM
Compress data
Autonomous car AI reduces accidents by:
Recognizing patterns of pedestrians and traffic
Random driving
FSM only
Timer automation
Security cameras using AI to detect intruders segment images to:
Identify moving objects accurately
Compress videos
Encrypt files
Store memory
Medical imaging AI detects tumors using:
Semantic segmentation (pixel-level labeling)
Random selection
FSM only
Timer automation
Self-driving cars use computer vision to:
Detect lanes, traffic signs, and obstacles
Compress videos
Store images only
BFS only
Face recognition in smartphones involves:
Feature extraction + pattern matching
FSM only
Timer automation
Random scanning
Drones inspecting fields use segmentation to:
Identify healthy vs. diseased crops
Randomly fly
FSM only
BFS
Edge detection helps AI:
Detect object boundaries in images
Compress images only
Encrypt outputs
Reduce FSM states
Augmented reality apps use:
Computer vision for overlaying virtual objects on real scenes
Random graphics
Timer display
FSM only
Retail stores use AI cameras to detect customer movement patterns. This is:
Pattern recognition + computer vision
FSM only
Random tracking
Timer automation
Autonomous robots sort objects using:
Vision-based feature detection
FSM only
BFS only
Timer automation
Image data augmentation helps AI:
Learn robustly from varied images
Reduce data
Compress images
Encrypt files
Dartmouth Conference (1956) led to:
Birth of AI research as a formal field
Invention of calculators
Robotics only
Neural networks only
AI ethics in facial recognition is important because:
Biased datasets can lead to unfair arrests or misidentification
Computers slow down
Memory is reduced
Random errors
Copyright concerns arise when AI-generated art:
Uses human-created images without permission
Compresses files
Stores features
Randomly generates images
Autonomous delivery drones must be accountable for accidents. This reflects:
Ethical AI principles
FSM rules
Bias in recruitment AI occurs when:
Training data reflects historical discrimination
Data is encrypted
BFS fails
Features are compressed
Fair AI in healthcare ensures:
Equal access and correct diagnosis for all patients
Faster outputs only
Random suggestions
FSM only
AI in autonomous cars must balance:
Safety, fairness, and decision transparency
FSM only
Memory usage
Timer rules
ELIZA chatbot (1960s) demonstrated:
Early natural language interaction with rule-based responses
Deep learning
Autonomous reasoning
Random outputs
AI winters occurred historically due to:
Overpromised results, limited hardware, and slow progress
FSM failures
Random events
Timer automation
Historical AI systems focused on:
Rule-based problem solving and logical reasoning
Deep learning only
Random movement
Timer automation
AI transparency means:
Users understand how decisions are made
Systems are invisible
Outputs are encrypted
FSM states are minimized
Ethical AI in finance ensures:
A. Lending decisions are fair and unbiased
B. Random approvals
C. Timer-based automation
D. BFS only
Reinforcement learning in real-life gaming apps allows AI to:
Learn optimal strategies via trial and error
Follow fixed rules only
Use FSM
Random moves
AI in social media recommends content to maximize engagement. Ethical concern:
Filter bubbles and bias amplification
BFS memory
FSM states
Random display
AI fairness means:
