Question 41
Which probability rule is fundamental for updating the belief or probability of a hypothesis given observed evidence?
Bayes' Theorem computes posterior probability P(H|E) using prior probability P(H) and likelihood P(E|H).
Which probability rule is fundamental for updating the belief or probability of a hypothesis given observed evidence?
Bayes' Theorem computes posterior probability P(H|E) using prior probability P(H) and likelihood P(E|H).
In Naive Bayes classification, what core assumption is made regarding features given the class label?
Naive Bayes assumes all predictor features are conditionally independent of each other given the class variable.
Which ensemble machine learning algorithm builds multiple decision trees on bootstrap dataset samples and averages their predictions?
Random Forest combines Bootstrap Aggregating (Bagging) with random feature selection across individual decision trees.
In AdaBoost (Adaptive Boosting), how are sample weights adjusted after each weak classifier is trained?
AdaBoost increases weights for misclassified training examples so subsequent weak learners focus on hard-to-classify samples.
Which dimensionality reduction technique transforms correlated features into a smaller set of uncorrelated orthogonal variables?
PCA projects data onto eigenvectors of the covariance matrix, maximizing variance along orthogonal principal axes.