Data Science with Python Certification Data Science with Python Natural Language Processing 3

Question 1

What type of neural network architecture is most commonly used for sequential NLP tasks such as text generation and machine translation?

Accepted Answer

Recurrent Neural Network (RNN) / LSTM

Answer

Convolutional Neural Network (CNN)

Answer

Feedforward Multilayer Perceptron

Answer

Autoencoder

Question 2

In the context of NLP, what is a word embedding?

Accepted Answer

A dense continuous vector representation that captures semantic relationships between words

Answer

A one-hot encoded vector where each dimension represents a vocabulary word

Answer

A statistical count of how often a word appears in a document

Answer

A binary feature indicating presence or absence of a word

Question 3

Which Python library provides pre-trained Word2Vec and FastText models for loading and using word embeddings?

Accepted Answer

Gensim

Answer

NLTK

Answer

spaCy

Answer

TextBlob

Question 4

What is the output of a text classification model trained with scikit-learn's `MultinomialNB` on a TF-IDF matrix?

Accepted Answer

A predicted class label for each document

Answer

A continuous regression value for each document

Answer

A cluster assignment for each document

Answer

A ranked list of similar documents

Question 5

What does the `similarity()` method return when called between two spaCy `Doc` objects?

Accepted Answer

A cosine similarity score between 0 and 1 based on their vector representations

Answer

The edit distance (Levenshtein distance) between the two texts

Answer

A list of common named entities found in both documents

Answer

The Jaccard similarity of the two documents' token sets

Question 6

In TextBlob, what does `TextBlob(text).sentiment.polarity` return?

Accepted Answer

A float between -1.0 (negative) and 1.0 (positive) representing sentiment

Answer

An integer count of positive words minus negative words

Answer

A string label such as 'positive', 'neutral', or 'negative'

Answer

The subjectivity score of the text on a 0–100 scale

Question 7

Which evaluation metric is most appropriate for a text classification task where class distribution is highly imbalanced?

Accepted Answer

F1-score (macro or weighted)

Answer

Accuracy

Answer

Mean Squared Error

Answer

Perplexity