Multiple choice questions
1. Forward state space search is also called ------------------------------
a) Sequential planning
b) Regression planning
c) Progression planning
d)None of the above
Ans: c) Progression planning
2. Which of the following is/are the components of the partial order planning.
a) Initial stage
b) Goal
c) Causal Links
d) Bindings
ANS: Goal, Causal Links, Bindings
3. Which of the following statements is/are correct?
a. P(a | b, c) + P(¬a | b, c) = 1
b. P(a | b, c) + P(a | ¬b, ¬c) = 1
c. P(a | b, ¬c) + P(a | b, c) = P(a | b)
d. P(a, b | c) = P(a | c).P(b | a, c)
4. . Consider the following Bayesian Network.
Statement 1: B is independent of E given evidence at C.
Statement 2: B is independent of A given evidence at E.
a. Both statements are correct.
b. Both statements are incorrect.
c. Statement 1 is correct and Statement 2 is incorrect.
d. Statement 1 is incorrect and Statement 1 is correct
5. The following statement is true for A Kalman Filter
a. Is an algorithm that takes data from known multiple sources and estimates unknown variables
b. Is an algorithm that takes data from known single source and estimates unknown variable
c. Is an algorithm that takes data from unknown multiple sources and estimates unknown variables
d. Is an algorithm that takes data from unknown single source and estimates unknown variable
ans: Is an algorithm that takes data from known multiple sources and estimates unknown variables
6. In HMM Where does the additional variables are added?
a)Probability over time
b) Reality model
c) Probability model
d) Temporal model
Ans: Temporal model
7._____________is the basic approach is the decision tree induction.
a)Procedural
b)Step by step
c)Greedy
d)Top-down
ANS: Top-down
8. Principle Component Analysis under------------------learning
a)Supervised Learning
b) Unsupervised Learning
c) Reinforcement Learning
d) Inductive Learning
ANS: Unsupervised Learning
9-Which technique is used for feature extraction for dimensionality decrementing?
a) PCA
b) KPCA
c) ICA
d) ALL
e) none
ANS:ALL
10. Which of the following sampling algorithms will ensure that no samples are discarded?
a. Rejection Sampling
b. Likelihood Sampling
c. Gibbs Sampling
Fill in the blanks
1. total-order planning is the opposite of Partial-order planning(True/False)_____
ANS: True
2 Define state space?
ANS:Set of all possible states of a given problem is
known as state space problem
• A state-space problem consists of
• a set of states;
• Initial or start state;
• a set of actions available to the agent in each state;
• an action function that, given a state and an action, returns a new state;
• a set of goal states, often specified as a Boolean function, goal(s), that is true when s is a goal state; and
3. There are two boxes A and B. Box A contains n red balls and 2 blue balls while box B contains 2 red balls and n blue
balls. One of the two boxes is selected at random and two balls are drawn from it without replacement. If both the
balls drawn are red, the probability of box A being selected is 6/7. Find the value of n-------------------------------
Ans: 4
4. Consider the following Bayesian Network with the given information. What is the value of P(¬a | d, e)---------------
P(a) = 0.001, P(b) = 0.002 P(c | a, b) = 0.95, P(c | a, ¬b) = 0.94, P(c | ¬a, b) = 0.29, P(c | ¬a, ¬b) = 0.001 P(d | c) = 0.9,
P(d | ¬c) = 0.05 P(e | c) = 0.7, P(e | ¬c) = 0.01
Ans: 0.716
5.List out the inference in temporal model techniques---------------------------------------------
ANS: filtering, smoothing, prediction
6. What are the classical approaches for probabilistic reasoning over a time.---------------------------
Ans: Hidden markov model,kalman filters and dynamic basien networks.
7. Another name of Dynamic Bayesian Network (DBN)---------------------------------------
Ans: Two-Time slice BN (2TBN)
8.Full form of CHAID decision tree algorithm-------------------------------------------
ANS: Chi-squared interaction Detector
9. Any three dis advantages of supervised learning
o Supervised learning models are not suitable for handling the complex tasks.
o Supervised learning cannot predict the correct output if the test data is different from the
training dataset.
o Training required lots of computation times.
o In supervised learning, we need enough knowledge about the classes of object.
10. What are the types of Induction learning algorithm?
K nearest neighbour,
locally weighted regression and
Case-based reasoning methods
