NLP

Question 1

Corpus

Answer 1

A collection of speech data. Used for training NLP models

Question 2

Word Sense Disambiguation

Answer 2

The process of identifying the correct meaning of a word, when multiple meanings can be interpreted.

Question 3

BLEU

Answer 3

Bi-Lingual Evaluation Understudy.
Metric used to measure the quality of a machine translation, compared to reference translations.

Question 4

ROUGE

Answer 4

A metric used to measure the quality of machine summaries, compared to reference summaries.

Question 5

Hidden Markov Model (HMM)

Answer 5

Statistical model used for part-of-speech tagging and speech recognition

Question 6

Part-of-Speech tagging

Answer 6

Tags words in a sentence according to noun, verb, adverb etc.

Question 7

Transfer Learning

Answer 7

Applying a model trained on one task to a different task.

Question 8

N-Gram

Answer 8

A sequence of N continuous items from a text of speech.

Question 9

Lemmatization

Answer 9

Reducing words to their base form, undoing any conjugations. Similar to Stemming.

Question 10

Stemming

Answer 10

Reducing words to their stem, undoing any conjugations. Similar to Lemmatization.

Question 11

Named Entity Recognition

Answer 11

Categorising names into predefined groups.

Question 12

Co-reference Resolution

Answer 12

Identifying which words in a text refer to the same entity.

Question 13

Stop Word

Answer 13

A commonly used word which does not contribute to a texts content.

Question 14

Word Embeddings

Answer 14

Vectorisation of words. Similar words are mapped to nearby vectors in vector space.

Question 15

Word2Vec

Answer 15

Word vectorisation method.

Question 16

GloVe

Answer 16

Word vectorisation method.

Question 17

BERT

Answer 17

Word vectorisation method.

Question 18

Bag-of-Words

Answer 18

Method for representing a set of words, without regard to order or grammar.

Question 19

TF-IDF

Answer 19

Term Frequency - Inverse Document Frequency
Metric measuring how important a word is to a document in a corpus, relative to its frequency in the rest of the corpus.

Question 20

Latent Semantic Analysis

Answer 20

Analyses relationships between words in a document corpus to discover semantic structures.

Question 21

Latent Dirichlet Allocation

Answer 21

Generative probabilistic model identifying topics in a document corpus.

Question 22

Generative Probabilistic Model

Question 23

Perplexity

Answer 23

Metric measuring how well a model predicts a sample. Lower perplexity is better performing.

Question 24

Componential Semantics

Answer 24

Words represented by sets of semantic components which together describe the meaning of the word.

Question 25

Distributional Semantics

Answer 25

Describing words according to the contexts they appear in.

Question 26

Thematic Distance

Answer 26

Metric measuring the similarity of words based on the angle between their vectors.

Question 27

Saltonian vector

Answer 27

Binary vector representation of a word. Zero everywhere except the index of the word in the corpuses complete wordlist.

Question 28

Vector size reduction

Question 29

Word error rate

Answer 29

A metric measuring the relative error between a generated text and a reference text.
S + D + I / N
S: Substitutions
D: Deletions
I: Insertions
N: Number of words in reference text

Question 30

Connectionist Temporal Classification

Answer 30

A method for end-to-end Automatic Speech Recognition.

Question 31

Attention-based Encoder-Decoder Models

Answer 31

A method for end-to-end Automatic Speech Recognition

Question 32

Transducer Models (RNN-T)

Answer 32

A method for end-to-end Automatic Speech Recognition