Recall & Review
beginner
What is the main idea behind the Naive Bayes algorithm in text classification?
Naive Bayes assumes that the presence of each word in a text is independent of the others and uses Bayes' theorem to calculate the probability that the text belongs to a certain category.
Click to reveal answer
beginner
Why is Naive Bayes called 'naive'?
Because it assumes that all features (words) are independent of each other, which is a simplification that is often not true in real language but still works well in practice.
Click to reveal answer
intermediate
What is the role of prior probability in Naive Bayes for text?
The prior probability represents how common each category is before seeing the text, helping the model to balance predictions based on category frequency.
Click to reveal answer
intermediate
How does Naive Bayes handle words that do not appear in the training data for a category?
It uses smoothing techniques like Laplace smoothing to assign a small non-zero probability to unseen words, preventing zero probability issues.
Click to reveal answer
beginner
What metric is commonly used to evaluate the performance of a Naive Bayes text classifier?
Accuracy is commonly used, which measures the percentage of correctly classified texts out of all texts tested.
Click to reveal answer
What assumption does Naive Bayes make about words in a text?
✗ Incorrect
Naive Bayes assumes that each word's presence is independent of others, simplifying calculations.
What does Laplace smoothing help with in Naive Bayes?
✗ Incorrect
Laplace smoothing assigns small probabilities to words not seen in training to avoid zero probabilities.
Which formula is central to Naive Bayes classification?
✗ Incorrect
Naive Bayes uses Bayes' theorem to calculate the probability of categories given the text.
In text classification, what does the 'prior' represent?
✗ Incorrect
The prior is the probability of each category before considering the text's words.
Which metric best shows how well a Naive Bayes text classifier works?
✗ Incorrect
Accuracy measures the percentage of correct predictions, showing model performance.
Explain how Naive Bayes uses word probabilities to classify a text.
Think about how the model combines word chances and category chances.
You got /5 concepts.
Describe why smoothing is important in Naive Bayes for text classification.
Consider what happens if a word never appeared in training for a category.
You got /4 concepts.