Natural Language Processing Week 4 NPTEL Assignment Answers 2025

NPTEL Natural Language Processing Week 4 Assignment Answers 2024

1. Baum-Welch algorithm is an example of

Options:
A) Forward-backward algorithm
B) Special case of the Expectation-maximisation algorithm
C) Both A and B
D) None

Answer :- C ✅
Explanation:

Baum-Welch is a specific implementation of the Expectation-Maximization (EM) algorithm for Hidden Markov Models (HMMs).
It uses the Forward-Backward algorithm in the expectation step.

Answer :- A ✅

3. In the question 2, the expected number of consecutive days of sunny weather is:

Options:
A) 2
B) 3
C) 4
D) 5

Answer :- d
Explanation:

Expected number of sunny days = 11−P(S→S)\frac{1}{1 – P(S \to S)}1−P(S→S)1
If transition probability from Sunny → Sunny = 0.8, then
11−0.8=5\frac{1}{1 – 0.8} = 51−0.81=5

4. Entropy calculation with known word probabilities (log base 10):

Vocabulary size = 1200,
200 stop words with p = 0.001 each

Options:
A) 2.079
B) 4.5084
C) 2.984
D) 3.0775

Answer :- d
Explanation:

Entropy H=−∑p(x)log⁡10p(x)H = – \sum p(x) \log_{10} p(x)H=−∑p(x)log10p(x)
For uniform distribution (best case), entropy = log⁡10(1200)≈3.079\log_{10}(1200) ≈ 3.079log10(1200)≈3.079
Since 200 stopwords have fixed p=0.001, remainder distributed across rest.
Maximum possible entropy with this constraint ≈ 3.0775

5. How many possible hidden state sequences for the sentence “Sachin Tendulkar is a great player”?

Tags given per word:

Options:
A) 4 × 3 × 3
B) 43°3
C) 2* × 23 × 23
D) 3 × 4⁴

Answer :- d
Explanation:

6. What are the space and time complexity of the Viterbi algorithm (K = states, N = time steps)?

Options:
A) KN, K°N
B) K^N, KN
C) K^N, K-N
D) KN, KN

Answer :- a
Explanation:

Answer :- b