0% found this document useful (0 votes)
6 views13 pages

Markov Networks: Key Concepts and Applications

The document is a quiz consisting of 30 questions related to Markov Networks and Bayesian Networks, focusing on concepts such as cliques, factors, and the differences between the two types of networks. It includes questions about the properties, advantages, and applications of each network type in various scenarios. The quiz aims to assess understanding of graphical models and their functionalities in probabilistic systems.

Uploaded by

guptasuryansh567
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
6 views13 pages

Markov Networks: Key Concepts and Applications

The document is a quiz consisting of 30 questions related to Markov Networks and Bayesian Networks, focusing on concepts such as cliques, factors, and the differences between the two types of networks. It includes questions about the properties, advantages, and applications of each network type in various scenarios. The quiz aims to assess understanding of graphical models and their functionalities in probabilistic systems.

Uploaded by

guptasuryansh567
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

TA-1 DL-2 12-08-

2024
kashyapsp@[Link] Switch account

Your email will be recorded when you submit


this form

TA-1 DL-2 12-08-2024

30 Questions : 30 Min

Which of the following is not true 1 point

about Markov Networks?

They are undirected

They can represent conditional


independencies

They require directionality between


nodes

They can be used to model


relationships in probabilistic systems
In the context of Markov Networks, 1 point

what does the term "clique" refer


to?

A subset of nodes where all nodes are


directly connected

A sequence of independent nodes

A single node

A node with no edges

What is a "Factor" in a Markov 1 point


Network?

A conditional probability

A function that captures the aAnity


between variables

An isolated node

A measurement of independence

Which concept cannot be directly 1 point

represented in a Bayesian Network


but can in a Markov Network?

Direct relationships

Symmetric dependencies

Conditional independencies

Causal relationships
What is a Perfect Map in the 1 point

context of graphical models?

A graph where nodes are perfectly


independent

A graph that exactly captures all


independencies of a distribution

A complete graph

A graph with no edges

What is the purpose of the 1 point

partition function Z in a Markov


Network?

To normalize the factors

To calculate probabilities

To measure the network size

To determine independence

Which of the following is an 1 point

example of a Markov Network?

Naive Bayes

Hidden Markov Model

Ising Model

Decision Tree
In a Markov Network, which of the 1 point

following properties is used to


determine local independencies?

Directed edges

Markov Blanket

Conditional probability

Latent variables

What is the key difference between 1 point

Bayesian Networks and Markov


Networks?

Bayesian Networks are undirected;


Markov Networks are directed

Bayesian Networks require fewer


parameters

Markov Networks do not have a


direction of inNuence between nodes

Bayesian Networks cannot model


conditional independence

Which of the following is true for a 1 point

clique in a Markov Network?

It is always a set of two nodes

It cannot be factorized

It is a maximal complete subgraph

It represents independent nodes


What type of factor is used in 1 point

Markov Networks to represent


interactions?

Univariate

Joint Probability

Conditional Probability

Potential Function

What is the purpose of the Gibbs 1 point

distribution in Markov Networks?

To estimate network size

To model causal relationships

To represent the joint distribution as a


product of factors

To eliminate conditional dependencies

In which of the following scenarios 1 point

would you prefer a Markov Network


over a Bayesian Network?

When relationships are directional

When modeling causal relationships

When dealing with symmetric


dependencies

When modeling temporal sequences


How are the weights in a Markov 1 point

Network typically determined?

Through causal inference

Through learning from data

Through random assignment

Through expert knowledge only

Which algorithm is commonly used 1 point

for inference in Markov Networks?

Naive Bayes

Belief Propagation

Gradient Descent

Expectation Maximization

Which type of problem is Markov 1 point

Networks particularly well-suited


for?

Time series forecasting

ClassiWcation with independent


features

Modeling symmetric relationships

Predicting future events


What is the role of the Markov 1 point

Blanket in a Markov Network?

It determines the network's topology

It isolates a node from the rest of the


network

It represents all the variables


independent of a given node

It includes the neighbors of a node,


making it independent of the rest

Which of the following is a 1 point

potential advantage of Markov


Networks over Bayesian
Networks?

Easier to understand

Better at handling large datasets

More Nexible in representing symmetric


relationships

Requires fewer parameters


What does the term "graphical 1 point

model" generally refer to?

A model based on graphs like trees or


networks

A model that represents data in visual


format

A statistical model that uses graphs to


represent dependencies

A model that visualizes mathematical


functions

In Bayesian Networks, what does a 1 point

directed edge signify?

Symmetric dependency

Temporal sequence

Causal inNuence

Mutual exclusivity

What is the importance of the 1 point

"parent" nodes in a Bayesian


Network?

They are always independent

They conditionally inNuence their child


nodes

They do not affect their child nodes

They are isolated nodes


In a Bayesian Network, what does 1 point

the term "non-descendants" refer


to?

Nodes that have no effect on the node


in question

Nodes that are not parents or children

Nodes that are causally related

Nodes that are isolated

What is a potential drawback of 1 point


using Bayesian Networks?

They are computationally expensive to


construct

They cannot model symmetric


relationships

They require a lot of data to be accurate

They are not interpretable


What is a "joint probability 1 point

distribution" in the context of


graphical models?

The probability distribution over all


variables in the network

The probability of two variables being


independent

The probability of a single variable

The distribution of factors across


cliques

What is the relationship between 1 point

cliques and factors in a Markov


Network?

Each factor corresponds to a clique in


the network

Factors are independent of cliques

Cliques determine the edges in a


Markov Network

Factors are only relevant in Bayesian


Networks
A company wants to model the 1 point

interactions between different


departments (Sales, Marketing, HR,
etc.) to understand how an issue in
one department might affect others
without assuming any directional
influence. Which model should they
use?

Bayesian Network

Decision Tree

Markov Network

Linear Regression

In a medical diagnosis system, you 1 point

need to predict whether a patient


has a certain disease based on
symptoms, where the presence of
one symptom increases the
likelihood of another. Which model
would be best to represent this
scenario?

Bayesian Network

Logistic Regression

Markov Network

Random Forest
You are designing a diagnostic 1 point

system for a car's engine, where a


fault in one part may directly cause
faults in other connected parts.
Which type of network should you
use to model this causal
relationship?

Bayesian Network

Markov Network

Random Forest

Clustering

A recommendation engine needs 1 point

to suggest products to users based


on the combination of products they
have already purchased, with the
assumption that certain products
are more likely to be purchased
together. Which model best
captures these relationships?

Markov Network

Bayesian Network

Support Vector Machine

Linear Regression
In a gene regulatory network, 1 point

genes influence each other’s


expression, but not necessarily in a
directed manner. What model would
be best to capture the
interdependencies among these
genes?

Bayesian Network

Markov Network

Neural Network

Linear Regression

Back Submit Clear form

Never submit passwords through Google Forms.

This form was created inside of RKNEC. - Contact form


owner

Forms

You might also like