Bayesian Networks

Front

What is a Bayesian Network (BN)?

Back

A probabilistic graphical model that represents variables and their conditional dependencies using a directed acyclic graph (DAG).

Front

What are the components of a Bayesian Network?

Back

1. Nodes: Random variables (e.g., Disease, Symptom). 2. Edges: Conditional dependencies (cause → effect). 3. Conditional Probability Tables (CPT): Probability distribution for each node given its parents.

Front

What is the structure of the Rain–Sprinkler–Wet Grass network?

Back

Node 1: Rain, Node 2: Sprinkler, Node 3: Wet Grass.

Front

What are the dependencies in the Rain–Sprinkler–Wet Grass network?

Back

Rain → Wet Grass and Sprinkler → Wet Grass.

Front

What is the formula for joint probability in Bayesian Networks?

Back

P(X1, X2, …, Xn) = Π P(Xi | Parents(Xi)).

Front

What does the joint probability of a Bayesian Network represent?

Back

The probability of the whole network, which is the product of each node’s probability given its parents.

Front

What is an example of a Conditional Probability Table (CPT) for the Rain–Sprinkler–Wet Grass network?

Back

Rain | Sprinkler | P(WetGrass=Yes): Yes | Yes | 0.99, Yes | No | 0.90, No | Yes | 0.80, No | No | 0.00.

Front

What are some applications of Bayesian Networks?

Back

1. Medical Diagnosis → Disease → Symptoms. 2. Spam Filtering → Words → Spam probability. 3. Decision Support Systems → Risks and outcomes. 4. Robotics → Sensor fusion under uncertainty.

Front

What are the advantages of using Bayesian Networks?

Back

1. Handles uncertainty effectively. 2. Provides a graphical view of dependencies. 3. Easy to update beliefs with new evidence. 4. Supports reasoning about cause–effect relationships.