1. What is Machine Learning?
Answer: Machine Learning is a branch of AI that enables computers to learn from data and make predictions without being explicitly programmed.
2. What is Supervised Learning?
Answer: Learning using labeled data where the correct output is already known.
Example: Spam email detection.
3. What is Unsupervised Learning?
Answer: Learning from unlabeled data to find hidden patterns.
Example: Customer segmentation.
4. Difference Between Supervised and Unsupervised Learning?
SupervisedUnsupervisedUses labeled dataUses unlabeled dataPredicts outputFinds patternsExample: ClassificationExample: Clustering
5. What is Classification?
Answer: Predicting categories or classes.
Example: Spam or Not Spam.
6. What is Clustering?
Answer: Grouping similar data points together.
Example: Customer groups.
7. What is a Dataset?
Answer: A collection of related data used for training and testing models.
8. What is Training Data?
Answer: Data used to teach the model.
9. What is Testing Data?
Answer: Data used to check model performance.
10. What is Accuracy?
Answer: Percentage of correct predictions made by the model.
Naive Bayes
11. What is Naive Bayes?
Answer: A classification algorithm based on Bayes' Theorem.
12. Why is it called Naive?
Answer: Because it assumes all features are independent.
13. Where is Naive Bayes used?
Answer: Spam filtering, document classification, sentiment analysis.
KNN
14. What is KNN?
Answer: K-Nearest Neighbors is a classification algorithm that predicts using nearby data points.
15. What does K represent?
Answer: Number of nearest neighbors.
16. Why is K important?
Answer: It affects prediction accuracy.
17. Distance measure used in KNN?
Answer: Euclidean Distance.
PCA
18. What is PCA?
Answer: Principal Component Analysis reduces the number of features while keeping important information.
19. Why use PCA?
Answer: To reduce dimensionality and improve performance.
K-Means
20. What is K-Means?
Answer: An unsupervised clustering algorithm.
21. What is K in K-Means?
Answer: Number of clusters.
22. Steps of K-Means?
Answer:
Choose K
Select centroids
Assign points
Update centroids
Repeat
Random Forest
23. What is Random Forest?
Answer: A collection of multiple decision trees.
24. Why Random Forest?
Answer: More accurate than a single decision tree.
25. What is a Decision Tree?
Answer: A tree-like model used for decision making.
SVM
26. What is SVM?
Answer: Support Vector Machine is a classification algorithm that separates data using a boundary.
27. What is a Hyperplane?
Answer: A line or boundary that separates classes.
28. What are Support Vectors?
Answer: Data points closest to the boundary.
Logistic Regression
29. What is Logistic Regression?
Answer: A classification algorithm used to predict categories.
30. Difference Between Linear and Logistic Regression?
Linear: Predicts continuous values.
Logistic: Predicts categories.
31. What is Sigmoid Function?
Answer: A function that converts values into probabilities between 0 and 1.
Gradient Descent
32. What is Gradient Descent?
Answer: An optimization algorithm used to minimize errors.
33. What is Learning Rate?
Answer: Controls how much the model learns in each step.
34. What is an Epoch?
Answer: One complete pass through the training data.
Python Viva
35. What is a List?
Answer: Ordered and mutable collection.
36. What is a Tuple?
Answer: Ordered and immutable collection.
37. What is a Set?
Answer: Unordered collection of unique values.
38. What is a Dictionary?
Answer: Stores data as key-value pairs.
39. Difference Between List and Tuple?
List: Mutable
Tuple: Immutable
Prolog Viva
40. What is Prolog?
Answer: A logic programming language used in AI.
41. What is a Fact?
Answer: A statement that is always true.
Example: parent(john,mary).
42. What is a Rule?
Answer: Defines relationships using conditions.
43. What is a Query?
Answer: A question asked to Prolog.
5 One-Line Answers Examiners Frequently Ask
What is AI?
Artificial Intelligence is the simulation of human intelligence by machines.
What is Overfitting?
When a model learns training data too well and performs poorly on new data.
What is Underfitting?
When a model fails to learn patterns from data.
What is Feature Scaling?
Making features have a similar range of values.
What is Dimensionality Reduction?
Reducing the number of features while keeping important information.
These 43 questions are enough for a quick viva revision before your exam. 🚀📖