LDA with scikit-learn in NLP - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the LDA model from scikit-learn.

NLP

from sklearn.decomposition import [1]

Drag options to blanks, or click blank then click option'

ALatentDirichletAllocation

BPCA

CTruncatedSVD

DKMeans

Attempts:

3 left

💡 Hint

Common Mistakes

Importing PCA or KMeans instead of LatentDirichletAllocation.

Using TruncatedSVD which is for dimensionality reduction, not topic modeling.

The correct import for LDA in scikit-learn is LatentDirichletAllocation.

2fill in blank

medium

Complete the code to create an LDA model with 5 topics.

NLP

lda = LatentDirichletAllocation(n_components=[1], random_state=42)

Drag options to blanks, or click blank then click option'

A3

B10

C5

D1

Attempts:

3 left

💡 Hint

Common Mistakes

Choosing 1 topic which is too few.

Choosing 3 or 10 which are not the requested 5 topics.

We want 5 topics, so n_components=5 is correct.

3fill in blank

hard

Fix the error in the code to fit the LDA model on the document-term matrix named 'dtm'.

NLP

lda.fit([1])

Drag options to blanks, or click blank then click option'

Adocuments

Bdtm

Clda

Dvectorizer

Attempts:

3 left

💡 Hint

Common Mistakes

Passing raw documents instead of the document-term matrix.

Passing the vectorizer or the model itself.

The LDA model fits on the document-term matrix, which is dtm.

4fill in blank

hard

Fill both blanks to get the topic-word distribution and the top words for the first topic.

NLP

topic_word = lda.[1]_
top_words = [vectorizer.get_feature_names_out()[i] for i in topic_word[[2]].argsort()[-10:]]

Drag options to blanks, or click blank then click option'

Acomponents

Btransform

C0

D1

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'transform' which is a method, not an attribute.

Using index 1 which is the second topic, not the first.

The components_ attribute holds topic-word distributions. Index 0 selects the first topic.

5fill in blank

hard

Fill all three blanks to transform documents to topic distributions and print the topic distribution for the first document.

NLP

doc_topic_dist = lda.[1](dtm)
print(doc_topic_dist[[2]])
print(doc_topic_dist.shape[[3]])

Drag options to blanks, or click blank then click option'

Atransform

B0

Dfit_transform

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'fit_transform' which fits and transforms but is not always needed here.

Using wrong indices for document or shape.

transform converts documents to topic distributions. Index 0 selects the first document. Shape index 0 gives number of documents.