Back to Subject

Unit 2: Vector Space Models - Practice Quiz

INT344 — Natural Language Processing 50 Questions
0 Correct 0 Wrong 50 Left
0/50

1 What is the fundamental representation of a word in a Vector Space Model (VSM)?

A. A linked list
B. A vector of real numbers
C. A scalar integer
D. A binary tree structure

2 Which of the following is a primary advantage of dense word vectors over one-hot encoding?

A. They are easier to calculate manually
B. They capture semantic relationships
C. They use more memory
D. They are strictly binary

3 In the context of the Continuous Bag-of-Words (CBOW) model, what is the input to the neural network?

A. A random noise vector
B. The center word
C. The entire document
D. The context words

4 How is 'cosine similarity' calculated between two word vectors, A and B?

A. Cross product of A and B
B. Dot product of A and B divided by the product of their magnitudes
C. Sum of elements in A minus sum of elements in B
D. Euclidean distance between A and B

5 If two word vectors have a cosine similarity of 1, what does this imply?

A. The vectors point in exactly the same direction
B. The words represent opposite meanings
C. The vectors are orthogonal
D. The words are unrelated

6 Which architecture predicts the surrounding words given a center word?

A. Latent Dirichlet Allocation
B. Skip-gram
C. Principal Component Analysis
D. Continuous Bag-of-Words (CBOW)

7 What mathematical technique is commonly used to visualize high-dimensional word vectors in two dimensions?

A. Principal Component Analysis (PCA)
B. Logistic Regression
C. Linear Regression
D. Fourier Transform

8 In vector arithmetic for analogies, what result is expected for vector('King') - vector('Man') + vector('Woman')?

A. Queen
B. Prince
C. Monarch
D. Princess

9 What is the role of the 'window size' hyperparameter in CBOW?

A. It determines the number of epochs
B. It determines the number of dimensions in the vector
C. It sets the learning rate
D. It defines how many neighbors to consider as context

10 Why are vector space models useful for information retrieval/document search?

A. They eliminate the need for indexing
B. They work best with images
C. They allow exact keyword matching only
D. They can match queries to documents based on semantic similarity

11 When transforming word vectors from one language to another (e.g., English to French) using a linear mapping, what are we trying to learn?

A. A clustering algorithm
B. A binary classifier
C. A decision tree
D. A rotation matrix

12 In PCA, the first principal component is the direction that maximizes what?

A. The number of clusters
B. The error rate
C. The cosine similarity
D. The variance of the data

13 Which of the following best describes the 'bag-of-words' model assumption?

A. Grammar rules are strictly enforced
B. Dependencies between words are preserved
C. Word order is ignored, only frequency counts matter
D. Word order is critical for meaning

14 In the CBOW model, how are the input context vectors usually handled before passing to the hidden layer?

A. They are multiplied
B. Only the first word is used
C. They are averaged or summed
D. They are concatenated

15 What does a cosine similarity of 0 indicate between two word vectors?

A. They are identical
B. They are orthogonal (unrelated)
C. One is a scalar multiple of the other
D. They are opposite

16 Which loss function is typically minimized when aligning two vector spaces (X and Y) via a transformation matrix R?

A. Cross-entropy loss
B. Accuracy score
C. Hinge loss
D. Frobenius norm of (XR - Y)

17 Deep Learning vector models like Word2Vec are often referred to as:

A. Sparse embeddings
B. Hierarchical clusters
C. Prediction-based embeddings
D. Count-based matrices

18 If 'Apple' and 'Pear' are close in vector space, this indicates:

A. Semantic similarity
B. Phonetic similarity
C. Syntactic similarity
D. Morphological similarity

19 When visualizing word vectors, why can't we simply plot the 300-dimensional vectors directly?

A. Human visual perception is limited to 2 or 3 dimensions
B. The vectors become binary
C. It would take too long to render
D. Computers cannot store 300 dimensions

20 Which of the following is NOT a benefit of using Vector Space Models in Machine Translation?

A. Improving alignment of synonyms
B. Handling rare words via similarity
C. Guaranteeing grammatically perfect sentences
D. Mapping entire languages without parallel corpora (unsupervised)

21 In a Word2Vec model, the dimension of the hidden layer corresponds to:

A. The number of training documents
B. The vocabulary size
C. The window size
D. The size of the word embedding vector

22 To perform document search using word vectors, how might one represent a whole document?

A. By taking the average (centroid) of all word vectors in the document
B. By using the vector of the first word only
C. By concatenating all vectors into one giant vector
D. By summing the ASCII values of characters

23 What is the 'Curse of Dimensionality' in the context of NLP?

A. The inability to use PCA
B. Having too few dimensions to represent meaning
C. Data becoming sparse and distance metrics becoming less meaningful in very high dimensions
D. The time it takes to train a model

24 Which algebraic structure is used to transform word vectors from a source language space to a target language space?

A. A tensor of rank 3
B. A vector
C. A transformation matrix
D. A scalar

25 In PCA, what are 'eigenvalues' used for?

A. To determine the direction of axes
B. To label the axes
C. To quantify the variance explained by each principal component
D. To calculate the dot product

26 Which word pair would likely have the highest Euclidean distance in a well-trained vector space?

A. Frog - Toad
B. Computer - Sandwich
C. Happy - Joyful
D. Car - Automobile

27 The output layer of a standard CBOW model typically uses which activation function to generate probabilities?

A. ReLU
B. Sigmoid
C. Tanh
D. Softmax

28 What is the main limitation of using Euclidean distance for word vectors compared to Cosine similarity?

A. It is sensitive to the magnitude (length) of the vectors
B. It only works in 2D
C. It cannot handle negative numbers
D. It is computationally harder

29 Which concept explains why 'Paris' is to 'France' as 'Tokyo' is to 'Japan' in vector space?

A. Linear substructures / Parallelism
B. Singular Value Decomposition
C. Orthogonality
D. One-hot encoding

30 How does PCA reduce dimensions?

A. By deleting the last 50 columns of data
B. By averaging all data points to zero
C. By removing words with fewer than 3 letters
D. By projecting data onto new axes that minimize information loss

31 In the context of relationships between words, 'distributional semantics' suggests that:

A. Words are defined by their dictionary definitions
B. Words are defined by their spelling
C. Words are unrelated entities
D. Words that appear in similar contexts have similar meanings

32 When training CBOW, what is the 'target'?

A. The part of speech
B. The center word
C. The next sentence
D. The sentiment of the sentence

33 What happens to the vectors of synonyms (e.g., 'huge' and 'enormous') during training?

A. They move infinitely far apart
B. They become orthogonal
C. One replaces the other
D. They move closer together

34 If you want to visualize a 1000-word subset of your vocabulary using PCA, what is the shape of the input matrix?

A. 1000 x 2
B. Dimension_of_Embedding x 1000
C. 2 x 2
D. 1000 x Dimension_of_Embedding

35 In cross-lingual information retrieval, query translation can be achieved by:

A. Multiplying the query vector by a transformation matrix
B. Re-training the model from scratch
C. Ignoring the language difference
D. Using a dictionary lookup only

36 What is a 'context window'?

A. The time limit for training
B. The graphical user interface
C. The software used to view the code
D. The number of words before and after a target word

37 Which of the following is NOT a step in performing PCA?

A. Calculating the covariance matrix
B. Standardizing the data
C. Applying a Softmax function
D. Computing eigenvectors and eigenvalues

38 If a word vector has a magnitude of 1, it is called:

A. A complex vector
B. A normalized vector
C. A sparse vector
D. A binary vector

39 Which approach is generally faster to train: CBOW or Skip-gram?

A. They are exactly the same
B. Neither is trainable
C. Skip-gram
D. CBOW

40 To capture dependencies between words that are far apart in a sentence, one should:

A. Set window size to 0
B. Increase the window size
C. Decrease the window size
D. Use one-hot encoding

41 The 'Manifold Hypothesis' in NLP suggests that:

A. High-dimensional language data lies on a lower-dimensional manifold
B. Vectors must be 3D
C. Language is flat
D. All words are equidistant

42 When performing vector arithmetic for 'Paris - France + Italy', the result is likely closest to:

A. London
B. Rome
C. Pizza
D. Germany

43 What is the dimensionality of the transformation matrix R used to map a source space of dimension D to a target space of dimension D?

A. D x 1
B. D x D
C. 2D x 2D
D. 1 x D

44 Which vector operation is primarily used to measure the relevance of a document to a search query in VSM?

A. Scalar Multiplication
B. Vector Subtraction
C. Cosine Similarity
D. Vector Addition

45 Sparse vectors (like Bag-of-Words) are characterized by:

A. Negative numbers only
B. Mostly zero values
C. Mostly non-zero values
D. Complex numbers

46 Word embeddings capture which type of relationships?

A. Both syntactic and semantic
B. Only syntactic
C. Neither
D. Only semantic

47 Before applying PCA, it is standard practice to:

A. Square the data
B. Invert the data
C. Randomize the data
D. Mean-center the data

48 In the analogy 'A is to B as C is to D', which equation represents the relationship in vector space?

A. B / A = D / C
B. B A = D C
C. B - A = D - C
D. B + A = D + C

49 Why might we use PCA on word vectors before performing clustering?

A. To increase the number of dimensions
B. To remove noise and reduce computational cost
C. To convert vectors to text
D. To translate the language

50 Which technique allows checking if the transformation matrix between two languages is accurate?

A. Measuring the vector length
B. Checking the accuracy of translation on a hold-out dictionary
C. Checking if the matrix is square
D. Calculating the determinant
Previous MCQ
Unit 1
Next MCQ
Unit 3