Why text classification categorizes documents in NLP - Challenge Your Understanding

Challenge - 5 Problems

🧠 Conceptual

intermediate

Purpose of Text Classification

Why do we use text classification in machine learning?

A To assign predefined categories to text documents based on their content

B To generate new text documents from existing ones

C To translate text documents from one language to another

D To compress text documents to save storage space

💡 Hint

Think about what it means to sort or label text automatically.

Text classification is used to automatically assign labels or categories to text based on its content, helping organize and understand large amounts of text data.

❓ Model Choice

intermediate

Choosing a Model for Text Classification

Which model is commonly used for text classification tasks?

A Linear Regression

B Convolutional Neural Network (CNN)

C K-Means Clustering

D Principal Component Analysis (PCA)

💡 Hint

Look for a model that can learn patterns in text sequences.

CNNs can capture local patterns in text and are effective for text classification tasks. K-Means is unsupervised clustering, Linear Regression is for continuous outputs, and PCA is for dimensionality reduction.

❓ Metrics

advanced

Evaluating Text Classification Accuracy

You trained a text classification model and got these results on test data: 80 true positives, 10 false positives, 15 false negatives, and 95 true negatives. What is the precision of the model?

A 0.80

B 0.84

C 0.75

D 0.89

💡 Hint

Precision = True Positives / (True Positives + False Positives)

Precision measures how many predicted positive cases are actually positive. Here, precision = 80 / (80 + 10) = 0.89.

🔧 Debug

advanced

Debugging Text Classification Code Output

What error will this code produce? ```python from sklearn.feature_extraction.text import CountVectorizer from sklearn.naive_bayes import MultinomialNB texts = ['spam message', 'ham message'] labels = [1, 0] vectorizer = CountVectorizer() X = vectorizer.fit_transform(texts) model = MultinomialNB() model.fit(X, labels) pred = model.predict(['spam message']) print(pred) ```

A TypeError: Expected array-like, got string instead

B [1]

C ValueError: Input contains NaN

D IndexError: list index out of range

💡 Hint

Check the input type for the predict method.

The predict method expects a vectorized input, not a raw string. Passing a string causes a TypeError.

🧠 Conceptual

expert

Why Text Classification Uses Predefined Categories

Why does text classification require predefined categories before training?

A Because categories are only used after training to evaluate the model

B Because the model generates new categories during training automatically

C Because the model learns to assign input texts to these known categories based on examples

D Because categories are irrelevant and not used in text classification

💡 Hint

Think about supervised learning and how labels guide the model.

Text classification is a supervised task where the model learns from labeled examples to assign new texts to the same predefined categories.