Sentiment analysis pipeline in NLP - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the necessary library for text vectorization.

NLP

from sklearn.feature_extraction.text import [1]

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

ATfidfVectorizer

BLabelEncoder

CCountVectorizer

Dtrain_test_split

Attempts:

3 left

💡 Hint

Common Mistakes

Choosing CountVectorizer which counts word occurrences but does not weight them.

Using LabelEncoder which is for labels, not text features.

The TfidfVectorizer converts text data into numerical features based on term frequency-inverse document frequency, which is commonly used in sentiment analysis.

2fill in blank

medium

Complete the code to split the dataset into training and testing sets.

NLP

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=[1], random_state=42)

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

A0.1

B0.2

C0.5

D1.0

Attempts:

3 left

💡 Hint

Common Mistakes

Using 1.0 which means all data is test set.

Using 0.5 which splits data evenly but is less common.

A common split for testing is 20% of the data, so test_size=0.2 is used here.

3fill in blank

hard

Fix the error in the model training code by completing the missing method.

NLP

model = LogisticRegression()
model.[1](X_train, y_train)

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

Apredict

Bscore

Ctransform

Dfit

Attempts:

3 left

💡 Hint

Common Mistakes

Using predict before training causes errors.

Using transform is for feature transformers, not models.

The fit method trains the model on the training data.

4fill in blank

hard

Complete the code to create a dictionary of predictions and calculate accuracy.

NLP

predictions = model.[1](X_test)
accuracy = accuracy_score(y_test, predictions,)

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

Apredict

B)

C,

Dpredict_proba

Attempts:

3 left

💡 Hint

Common Mistakes

Using predict_proba which returns probabilities, not labels.

Missing the comma between arguments in accuracy_score.

Use predict to get predicted labels, then separate arguments with a comma in accuracy_score.

5fill in blank

hard

Fill all three blanks to build a simple sentiment analysis pipeline.

NLP

from sklearn.pipeline import Pipeline
pipeline = Pipeline([
    ('vectorizer', [1]()),
    ('classifier', [2]())
])
pipeline.[3](X_train, y_train)

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

ATfidfVectorizer

BLogisticRegression

Cfit

DCountVectorizer

Attempts:

3 left

💡 Hint

Common Mistakes

Using CountVectorizer instead of TfidfVectorizer which weights terms.

Calling predict instead of fit to train the pipeline.

The pipeline uses TfidfVectorizer to convert text, LogisticRegression as the model, and fit to train.