Natural Language Processing
CSE · Quiz · Summer 26
Paper 1 of 3By Abir Mia -231Switch
Paper 1 of 3
Read the questions as text
Read off the PDF by AI, so it may have mistakes. The PDF is the original.
Daffodil International University
Department of Computer Science and Engineering
Faculty of Science & Information Technology
Class Test - 1, Summer-2026
Course Code: CSE432
Course Title: NLP
Time: 45 minutes
Full Marks: 15
Q1. Consider the following three documents:
D1: "data science is fun"
D2: "data science is hard"
D3: "football is fun"
(a) Compute the Cosine Similarity between:
• D1 and D2
• D1 and D3
(b) Which pair of documents is most similar according to Cosine Similarity?
Q2. Given the following corpus of 4 documents:
D1: "the cat sat on the mat"
D2: "the cat sat on the hat"
D3: "the dog sat on the log"
D4: "the bird flew over the mat"
(a) Calculate the TF-IDF score for "cat", "mat", and "sat" in document D1.
(b) Which word gets the highest TF-IDF score in D1?
Q3. A spam classifier is trained on the following dataset:
D1: "win free money now" -> Spam
D2: "free lottery win prize" -> Spam
D3: "meeting at noon today" -> Not Spam
D4: "lunch meeting tomorrow noon" -> Not Spam
D5: "win the lottery today" -> Spam
Test Document: "win free lunch"
(a) Calculate the Prior Probabilities.
(b) Calculate the Likelihood of each word in the test document for both classes using Laplace Smoothing.
(c) Which class does the classifier assign to the test document?
Q4. What is Maximum Likelihood Estimation (MLE) in the context of Bigram models?
Write the MLE formula for computing a Bigram probability and explain each component.