Data Science and Big Data Analytics (MA-462)

Which of the following best describes the Bag-of-Words (BoW) model?

• It encodes semantic meaning of words in low-dimensional vectors.

• It considers the order of words in a sentence.

• It converts text into fixed-length vectors based on word occurrence.

• It uses neural networks to learn word embeddings.

What is the primary role of the IDF (Inverse Document Frequency) component in TF-IDF?

• To increase the weight of frequently occurring terms in all documents.

• To reduce the weight of commonly used terms and highlight rarer terms.

• To normalize term frequencies across sentences.

• To generate low-dimensional word vectors.

In the context of Word2Vec, how is the vector for a word like "hat" learned?

• Through frequency counting in documents.

• By calculating TF and IDF values.

• By using a neural network that learns to predict context words (Skip-Gram) or the target word (CBOW).

• By averaging character-level embeddings.

Why is Word2Vec considered more "context-aware" than BoW or TF-IDF?

• Because it stores the exact position of each word in the sentence.

• Because it encodes words as binary strings.

• Because it learns word meanings based on surrounding words during training.

• Because it includes part-of-speech tags in the representation.

In the ID3 algorithm, which metric is used to select the best attribute at each node?

• Mean Squared Error

• Gini Index

• Entropy

• Information Gain

If an attribute results in zero information gain, what does it imply?

• It has the highest entropy

• It is the best attribute for splitting

• It provides no useful information for classification

• It must be the root node

What effect does pruning have on bias and variance?

• Decreases both bias and variance

• Increases bias, decreases variance

• Decreases bias, increases variance

• Has no impact on bias or variance

Which of the following is a limitation of the ID3 algorithm?

• Handles continuous data efficiently

• Supports pruning internally

• Biased towards attributes with many values

• Uses Gini Index for splitting

Which of the following is a common web crawling technique used by large-scale search engines?

• Random Forest Crawling

• Focused Crawling

• Recursive Machine Learning

• Binary Backtracking

How does a breadth-first crawling strategy operate?

• It prioritizes the most recently updated pages

• It follows the most linked pages first

• It visits pages level by level from a seed URL

• It selects pages based on semantic similarity