Machine Learning (IT-802 (A)) - Important Questions

Explain the differences between supervised, unsupervised, semi-supervised and reinforcement learning with practical examples for each.

Compare finite and infinite hypothesis spaces in machine learning. What are the implications of each on model complexity and generalization?

Describe the key concepts of Support Vector Machines for linear classification and how the margin is optimized.

Explain the role of activation functions in a multilayer perceptron. How do they enable the learning of nonlinear patterns?

Differentiate between linear regression, multiple linear regression, and logistic regression. Provide examples where each would be appropriate.

Explain the ID3 algorithm for building decision trees. How does it use entropy and information gain?

Discuss the role of meta-learners in stacking and how they contribute to the overall model performance.

What is the bagging technique in ensemble learning, and how does it reduce variance in predictions?

Explain hierarchical clustering and compare AGNES and DIANA methods, including their advantages and limitations.

Compare PCA and Locally Linear Embedding (LLE). When would you prefer LLE over PCA?

Describe Gaussian Mixture Models (GMM) and how they model data clusters as probabilistic distributions. What makes GMMs suitable for overlapping clusters?

Suppose that the data mining task is to cluster the following eight points (with (x, y) representing location) into three clusters: A1(2,10), A2(2,5), A3(8,4), B1(5,8), B2(7,5), B3(6,4), C1(1,2), C2(4,9). Apply K-means clustering, show initial assignment and update of cluster centres for two iterations.

What is frequent pattern mining and why is it important in market basket analysis?

Explain the process of constructing a Bayesian Belief Network, including defining nodes, edges and conditional probability tables.

State the assumptions of the Naive Bayes classifier and explain how they simplify probability computation.

Apply Bayes' theorem to compute posterior probability of drug use given a positive test with given false positive, false negative and prior rates.