Loading course title...

Loading assessment title...

Connecting to LMS... Progress: in progress

Assessment

1. What are the two main parts of a Bayesian network? A. A spreadsheet and a dashboard B. A neural network and a decision tree C. A directed acyclic graph and local probability distributions D. A database and a web server

2. What does a node normally represent in a Bayesian network? A. A random variable B. A guaranteed outcome C. A software function D. A physical cable

3. What is a posterior probability? A. A probability that must equal zero B. A graph edge C. A probability used before evidence D. A probability updated after considering evidence

4. What does a conditional probability table describe? A. Only the names of variables B. A variable's probability distribution for different parent states C. The visual position of each node D. The model's file format

5. What is diagnostic reasoning? A. Moving only from causes toward effects B. Deleting evidence from a model C. Reasoning from observed effects toward possible causes D. Proving that one variable causes another

6. What does explaining away describe? A. Removing every uncertain variable B. Ignoring base rates C. Converting a graph into a spreadsheet D. One confirmed cause making an alternative cause less likely for the same observed effect

7. What is parameter learning? A. Estimating probability values from data when the network structure is known B. Automatically proving causality C. Removing all graph edges D. Replacing evidence with assumptions

8. Which is the best Bayesian network workflow? A. Draw arbitrary arrows and treat the output as certain B. Define the question, select variables, build the graph, define probabilities, enter evidence, validate results, and document limitations C. Ignore model drift and missing variables D. Assume every edge proves causation