BCS602 – Machine Learning Solved Super Important Questions

Module 1

1] Explain the need for Machine Learning in modern organizations. What factors have contributed to its growing popularity?

2] What is Machine Learning? Explain how it differs from traditional programming. Discuss the process of learning in humans and machines with the help of examples.

3] Explain how Machine Learning is related to other major fields [or] Define Machine Learning. Explain its relationship to other fields with diagram

4] Explain different types of machine learning with a diagram

5] Explain the challenges of Machine learning. [or]Explain the major challenges faced in the implementation of Machine Learning systems.

6] Explain the Machine Learning Process model/ Data Mining Process. (with the help of CRISP-DM model.)

7] List and explain any four real-world applications of Machine Learning.

8] Define data. Explain the 6 V’s of Big Data. / Elements of Big Data

9] What is Big Data Analytics? Explain the types of data analytics and the Big Data Analytics Framework.

10] Explain data preprocessing with an example. / Explain data collection and data preprocessing in big data analytics. Also describe different data cleaning and normalization techniques.

11] What is Descriptive Statistics? Explain different types of data with examples.

Module 2

1] Explain the importance of Statistics and Probability in Machine Learning. Briefly describe different types of probability distributions used. /
[or] Explain continuous and discrete probability distributions

2] Apply and explain Principal Component Analysis algorithm for the given data points and prove that PCA works.

3] Apply Singular Value Decomposition (SVD) to the given matrix and explain the steps involved.

4] Explain the design of a learning system / Design a learning system for chess game

5] Apply the Find-S algorithm to the given training dataset and explain the hypothesis generation process. Mention the limitations of the Find-S Algorithm.

6] Explain Version Spaces and the Candidate Elimination Algorithm.
Apply the algorithm to the given dataset and show how the version space is derived.

7] Explain and apply Candidate Elimination Algorithm for the given dataset.

8] Explain Modeling in Machine Learning

Module 3

1] Explain k-Nearest Neighbor (k-NN) algorithm. Using the student dataset given below, classify a new instance (6.1, 40, 5) using k-NN with k = 3. Show all calculations.

2] Define Weighted k-Nearest Neighbor algorithm. Apply weighted KNN algorithm using the given dataset to classify the test set data (7.6, 60,8) where k=3

3] Explain Nearest Centroid Classifier. Apply it on the following dataset to predict the class of a test instance (6, 5):

x	y	Class
3	1	A
5	2	A
4	3	A
7	6	B
6	7	B
8	5	B

4] Explain why instance-based learning is called lazy learning. Compare instance-based learning and model-based learning.

5] Explain the different types of regression methods used in machine learning. Also, mention the limitations of regression methods.

6] Distinguish between:
  i. Locally Weighted Regression and Linear Regression
  ii. Multiple Linear Regression and Logistic Regression

7] Analyze decision tree learning with its structure, advantages, and disadvantages. Explain its Algorithm with advantages and disadvantages

8] Explain the process of constructing a decision tree using ID3.
Make use of entropy and information gain to discover the root node for the
decision tree for the following dataset using ID3 algorithm. (Explain ID3)

9] Given a dataset with attributes such as CGPA, Practical Knowledge, and Communication Skills, explain how the C4.5 algorithm chooses the root node using Gain Ratio. (Explain C4.5)

10] Construct a regression tree using the following Table which consists of 10 data instances and 3 attributes ‘Assessment’, ‘Assignment’ and ‘Project’. The target attribute is the ‘Result’ which is a continuous attribute. (Explain Regression Tree)

Module 4

1] Explain Bayes Theorem, Maximum A Posteriori (MAP) Hypothesis (hMAP), and Maximum Likelihood (ML) Hypothesis (hML).
[or] Discuss the fundamentals of Bayes’ Theorem and describe how it is applied for classification using the Bayes model.
[or] Define Prior Probability. Explain Bayes Theorem, hML and hMAP with an example.

2] Explain the Naive Bayes algorithm. Mention its working steps and applications.

3] Explain Bayes Optimal Classifier and solve to find whether a patient is diagnosed as COVID Positive or COVID Negative using the table given below.

4] Analyze the student performance using Naive Bayes Algorithm for continuous attributes. Predict whether a student will get a job offer or not in the final year.

5] Explain the concept of Artificial Neurons. Describe the Simple model of an Artificial Neuron (McCulloch and Pitts Model.)

6] Define Activation Function. Explain different types of activation functions.

7] Explain the Perceptron Model and its Learning Algorithm.

8] Analyze different types of Artificial Neural Networks with a diagram.

9] Explain the Applications, Advantages, Disadvantages, and Challenges of Artificial Neural Networks (ANNs).

Module 5

1] Differentiate between Classification and Clustering.
Also explain Applications, Challenges, Advantages, and Disadvantages of Clustering.

2] Explain k-means clustering. List any two advantages and two disadvantages of it.

3] Apply K-Means Clustering Algorithm for the given data with initial seeds as objects 2 and 5:

Object	X-Coordinate	Y-Coordinate
1	2	4
2	4	6
3	6	8
4	10	4
5	12	4

4] Explain the Density-Based Methods (DBSCAN) and Grid-Based Approach (CLIQUE) in detail.

5] Explain Mean-Shift Clustering Algorithm

6] Determine characteristics, applications, and challenges of Reinforcement Learning.

7] Explain how reinforcement learning is different from supervised and unsupervised learning. Give suitable examples.

8] Analyze components of Reinforcement Learning with a diagram.

9] What is Q-learning? Explain how Q-values are updated. Also explain how SARSA is different from Q-learning.

10] Explain Model-Free Methods in reinforcement learning (Monte Carlo methods and Temporal Difference (TD) learning)