AMT 402- CONCEPTS IN NATURAL LANGUAGE PROCESSING

QUESTION BANK

MODULE I

1. Explain the main challenges of NLP.

2. Explain Heuristic-based NLP.
3. Explain any three applications of NLP.
4. Explain Machine Learning-based NLP
5. Explain the different approaches to solve NLP problems
6. What is a kernel function? What is the need for a kernel function? Can a kernel function be
replaced by an ordinary mapping function?
7. How is classification done by SVM on linearly separable data?
8. Illustrate the steps involved in classification in Naïve Bayes Classifier

MODULE II

1.Explain word concept word embeddings with an example?Explain word concept word embeddings
with an example?
2.What are the main differences between skip gram and continuous bag of words?
3.State Bayes’ Theorem.
4.Explain key stages of NLP Pipeline
5.Which techniques are used to convert text to numerical vectors?
6.How to solve text pre-processing
7.Explain the Feature Extraction techniques
8.Explain different Distributed Representation
9.Given the following data about documents and contents, use tf-idf document
scoring method to retrieve the document for the query Data Scientists
Doc 1: Ben studies about computers in Computer Lab.
Doc 2: Steve teaches at Brown University.
Doc 3: Data Scientists work on large datasets

10.Explain the stages of NLP pipeline

11.Explain Bag-of-ngrams model with an example.
 Module Ⅲ

1.The size of the corpus is 10,000,000 million documents,if we assume 0.3 million million
documents contain the term ‘cat’,then find TF-IDF score?
2.Differentiate Rule based classification and machine learning based classification?

3.Consider a document containing 100 words wherein the word cat appears 3
times. Now, assume we have 10 million documents and the word cat appears in
one thousand of these. Compute the normalized tf and the tf-idf and compare
them.
4.Differentiate Part-of-Speech(POS) tagging and Named Entity recognition(NER)
5.Explain Application of Text Classification.
6.Explain the pipeline for building text classification systems
7.Explain Sentiment Analysis with Logistic Regression.
8.Explain key stages of Information Extraction Pipeline
9.Explain Sentiment Analysis with SVM.
10.Explain Name Entity Recognition using Sequence Labelling with an example

11.Explain Bayesian Methods for Text Classification.

12.What is NER? How it works? How can we turn this structured problem into a sequence problem
like POS tagging, with one label per word?

Module Ⅳ

1. How can we used Relation Extraction via supervised learning?

2. How can machine learning be used to extract relevant information from customer reviews?
3. How is a Relation Analysis Systems evaluated?
4. Explain Vector space model for information retrieval
5. Consider these documents:
Doc 1 breakthrough drug for schizophrenia
Doc 2 new schizophrenia drug
Doc 3 new approach for treatment of schizophrenia
Doc 4 new hopes for schizophrenia patients
ⅰ)Draw the term-document incidence matrix for this document collection.
ⅱ)Draw the inverted index representation for this collection

6. Explain supervised approach to relation analysis. What are its limitations?

7. Given the following data about documents and contents

Create inverted index for 4 sample documents above?

8. Explain Term weighting and document scoring

9. Explain Supervised Learning Approaches to Relation Analysis

10.Explain inverted index in IR

For example, consider the following documents:
Document 1: The quick brown fox jumped over the lazy dog.
Document 2: The lazy dog slept in the sun.
Create an inverted index for these documents?

11. Explain Term weighting and Document scoring

12 consider three sentences and then compute their cosine similarity:

1. We went to the pizza place and you ate no pizza at all
2. I ate pizza with you yesterday at home
3. There’s no place like home

Module Ⅴ

1. Explain Evaluation Of Factoid answers?

2. What advantages does word alignment techniques can be exploited to do phrase alignment.
3. Explain evaluation metric for factoid question answering
4. What are the components that make up a noisy channel model of statistical Machine Translation?
5. Explain the phases of a factoid question-answering system
6. What are the components that make up a noisy channel model of statistical Machine
Translation?

7. Explain how is decoding done in a Phrase-based Statistical Machine Translation System?

8. Explain the concept of Mean Reciprocal Rank.

9. Explain a statistical algorithm for word alignment in Machine Translation.

10. Explain the features to be considered in passage revival of a factoid question

answering system
11. Explain the phases of a factoid question-answering system
12. Explain Decoding for Phrase-based Statistical Machine Translation with an
example.