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

RNNs and their gated variants (LSTMs, GRUs) maintain hidden state across time steps, making them well-suited for sequential text where word order matters.

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

Word embeddings (e.g., Word2Vec, GloVe) map words to dense low-dimensional vectors where semantically similar words are geometrically close, unlike sparse one-hot encodings.

Question 3

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

Accepted Answer

Gensim

Answer

Gensim specializes in topic modeling and word vector algorithms; its `KeyedVectors` API makes it straightforward to load pre-trained Word2Vec and FastText models.

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

`MultinomialNB` is a supervised classifier that assigns each document to one of the discrete class labels it was trained on (e.g., spam/not-spam, topic categories).

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

spaCy's `doc.similarity(other_doc)` computes the cosine similarity between the averaged word vectors of the two documents, returning a float between 0 and 1.

Data Science with Python Certification Practice Test

Data Science with Python Certification Practice Test

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