CS3491 - AIML Lab Viva Questions with
Answers (Regulation 2021)
Experiment 1: Depth First Search (DFS)
Q: What is DFS?
A: DFS (Depth First Search) is a graph traversal algorithm that explores as far as possible
along each branch before backtracking.
Q: Is DFS informed or uninformed search?
A: DFS is an uninformed search algorithm.
Q: What data structure is used in DFS?
A: Stack is used, either explicitly or through recursion.
Q: Where is DFS used in real-world applications?
A: Used in pathfinding, scheduling, and puzzle solving.
Experiment 2: Breadth First Search (BFS)
Q: What is BFS?
A: BFS (Breadth First Search) explores all nodes at the present depth before moving to the
next level.
Q: How does BFS differ from DFS?
A: BFS uses a queue and explores level-by-level, while DFS uses a stack and goes depth-first.
Q: What is the time complexity of BFS?
A: O(V + E), where V is vertices and E is edges.
Q: Where is BFS used in AI?
A: Used in shortest path finding, peer-to-peer networks, and web crawlers.
Experiment 3: Best First Search
Q: What is Best First Search?
A: Best First Search uses a heuristic to choose the next node that appears to be closest to the
goal.
Q: Is it an informed or uninformed search?
A: It is an informed search algorithm.
Q: What is the heuristic function in Best First Search?
A: Heuristic estimates the cost from the current node to the goal.
Experiment 4: A* Algorithm
Q: What is A* algorithm?
A: A* is an informed search algorithm that uses g(n) + h(n) to find the lowest cost path.
Q: What are g(n) and h(n) in A*?
A: g(n) is the cost from start to current node, h(n) is estimated cost to goal.
Q: Why is A* both complete and optimal?
A: If the heuristic is admissible, A* finds the shortest path and will always find a solution if
one exists.
Experiment 5: Support Vector Machine (SVM)
Q: What is supervised learning?
A: Learning from labeled data to map input to output.
Q: What is the purpose of SVM?
A: To classify data by finding the best separating hyperplane.
Q: How does SVM find the optimal hyperplane?
A: By maximizing the margin between different class data points.
Experiment 6: K-Means Clustering
Q: What is unsupervised learning?
A: Learning patterns from data without labeled output.
Q: What is K-Means clustering?
A: A clustering algorithm that partitions data into K groups.
Q: How does the K-Means algorithm work?
A: It assigns points to the nearest centroid and updates centroids until convergence.
Experiment 7: Naive Bayes
Q: What is the Bayes Theorem?
A: P(A|B) = P(B|A) * P(A) / P(B)
Q: Why is it called 'Naive'?
A: It assumes that all features are independent.
Q: What are the assumptions made in Naive Bayes?
A: Features contribute independently to the outcome.
Experiment 8: Decision Tree
Q: What is a Decision Tree?
A: A flowchart-like structure used for classification and regression.
Q: What are entropy and information gain?
A: Entropy measures impurity; information gain measures the reduction in entropy.
Q: How is the best attribute selected in a tree?
A: Using maximum information gain or gain ratio.
Experiment 9: K-Nearest Neighbors (KNN)
Q: What is KNN algorithm?
A: A lazy learning algorithm that classifies based on the majority of K nearest points.
Q: How does KNN classify a new data point?
A: By measuring distance to neighbors and taking a majority vote.
Q: What distance metrics are used in KNN?
A: Euclidean, Manhattan, or Minkowski distances.
Experiment 10: Linear Regression
Q: What is Linear Regression?
A: A method to model the relationship between a dependent and one or more independent
variables.
Q: What is the cost function in Linear Regression?
A: Mean Squared Error (MSE).
Q: What is gradient descent?
A: An optimization technique to minimize the cost function.
Experiment 11: Logistic Regression
Q: What is the difference between linear and logistic regression?
A: Linear predicts continuous output; logistic predicts probabilities for classification.
Q: What is the sigmoid function?
A: A function that maps input to range between 0 and 1.
Q: Where is logistic regression used?
A: Used in binary classification problems like spam detection.
Experiment 12: Artificial Neural Networks (ANN)
Q: What is an artificial neural network (ANN)?
A: A network of interconnected nodes inspired by human brain to process data.
Q: What is backpropagation?
A: An algorithm to update weights in ANN using gradient descent.
Q: What are activation functions in ANN?
A: Functions like ReLU, Sigmoid, and Tanh used to introduce non-linearity.
