One of the technologies that regularly pops up on lists of what you should know (or learn) to increase your IT skills is NLP: Natural Language Processing. While this has been around for a while, the concepts and technology are being used more and more with each passing day.

How much to you know about Natural Language Processing and how well can you answer questions on its basic principles? What follows is a self-test of 25 questions based on the general concepts and topics related to the technology.

In all cases, pick the best answer(s) to each question. The answers appear at the end of the questions. Good luck!

1. Which of the following involves breaking a string of characters into discrete items (words, punctuation marks, etc.)?
A. Tokenization
B. Apportioning
C. Divvying
D. Partitioning

2. The statistical structure of a piece of text can be approximated/analyzed using what model that is now widely used in speech recognition, handwriting recognition, data compression, and spam recognition?
A. Wiese Paradigm
B. Buck Pattern
C. Truitt Prototype
D. Markov Model

3. Which of the following is a computer-readable collection of text or speech?
A. Corpus
B. Aggregation
C. Accrual
D. Accretion

4. What is the process of reducing derived words to their root form known as?
A. hulling
B. skinning
C. coring
D. stemming

5. The branch of linguistics that focuses on the way in which words are formed from morphemes is known as?
A. Phenology
B. Morphology
C. Phrasology
D. Linkology

6. Within NLP, which of the following is a word, or phrase, that occurs together with a noun and acts as a reference to that noun (or phrase)?
A. Copula
B. Adposition
C. Particle
D. Determiner

7. Which of the following is a parsed text corpus that annotates syntactic or semantic sentence structure?
A. Compendium
B. Treebank
C. Anthology
D. Assemblage

8. With Part of Speech (POS) tagging, each tagged token is known as a(n):
A. Tuple
B. Twig
C. Stalk
D. Trunk

9. A mathematical model of computation in which an abstract machine can be in one of a finite set of possibilities at a time is known as a(n):
A. Non-deterministic processer
B. Hierarchical state machine
C. Circumvesuviana
D. Finite-state automata

10. Which of the following is the interdisciplinary branch of linguistics that combines research from both psychology and linguistics?
A. Cognitive linguistics
B. Computational semantics
C. Conditional morphology
D. Qualified etymology

11. Which of the following is an approach that uses artificial neural networks to predict the likelihood of a sequence of words?
A. XRC
B. NMT
C. UGFI
D. IFC

12. A treebank that is not in semantic structure will be in what type of structure?
A. formal
B. conceptual
C. cognitive
D. syntactic

13. Which of the following involves the problem of resolving references to earlier or later items in a sentence, paragraph, or discourse?
A. Allan ruling
B. Joyce decree
C. Harrison conundrum
D. Anaphora resolution

14. The base, or dictionary form, of a word is known as a(n):
A. stub
B. lemma
C. root
D. radicle

15. Which part of NLP is useful in marketing for looking at social media posts, identifying polarity, and finding public opinion trends?
A. Semantic parsing
B. eWOM
C. Sentiment analysis
D. Semantic role labelling

Please visit GoCertify to attempt the remaining 10 questions of this quiz.



ANSWERS

1. A — Tokenization involves breaking a string of characters into discrete items (words, punctuation marks, etc.).
2. D — With the Markov Model, text can be approximated/analyzed. It is now widely used in speech recognition, handwriting recognition, data compression, and spam recognition.
3. A — A corpus is a computer-readable collection of text or speech.
4. D — The process of reducing derived words to their root form is known as stemming.
5. B — The branch of linguistics that focuses on the way in which words are formed from morphemes is known as Morphology.
6. D — Within NLP, a determiner is a word, or phrase, that occurs together with a noun and acts as a reference to that noun (or phrase).
7. B — A treebank is a parsed text corpus that annotates syntactic or semantic sentence structure.
8. A — With Part of Speech (POS) tagging, each tagged token is known as a tuple.
9. D — A mathematical model of computation in which an abstract machine can be in one of a finite set of possibilities at a time is known as a finite-state automata.
10. A — Cognitive linguistics is the interdisciplinary branch of linguistics that combines research from both psychology and linguistics.
11. B — NMT (Neural machine translation) is an approach that uses artificial neural networks to predict the likelihood of a sequence of words.
12. D — A treebank that is not in semantic structure will be in syntactic structure.
13. D — Anaphora resolution involves the problem of resolving references to earlier or later items in a sentence, paragraph, or discourse.
14. B — The base, or dictionary form, of a word is known as a lemma.
15. C — Sentiment analysis is useful in marketing for looking at social media posts, identifying polarity, and finding public opinion trends.