Week 2

Question 1

What is a Lexicon in NLP?

Answer 1

A lexicon is a collection of words, often with definitions, parts of speech, and other linguistic information used to analyze language.

Question 2

What are Stopwords?

Answer 2

Stopwords are common words (e.g., “the”, “and”, “is”) that are often removed in NLP tasks because they add little value to the meaning of a text.

Question 3

Define WordNet in NLP

Answer 3

WordNet is a semantic dictionary of English, grouping words into synsets (synonym sets) with definitions and usage examples

Question 4

What is a Synset in WordNet?

Answer 4

A synset is a set of synonyms that represent one concept in WordNet, providing a richer structure than traditional dictionaries.

Question 5

Describe the difference between Stemming and Lemmatization.

Answer 5

Stemming removes affixes to find the word stem (e.g., “running” to “run”), while lemmatization reduces a word to its dictionary form (lemma), considering context.

Question 6

What is Tokenization?

Answer 6

Tokenization is the process of breaking text into individual tokens (words or punctuation), which simplifies text analysis.

Question 7

Why is Named Entity Recognition (NER) important in NLP?

Answer 7

NER identifies and classifies entities in text, like names, locations, and dates, helping to extract specific information.

Question 8

What are N-Grams in language modeling?

Answer 8

N-Grams are sequences of N words used to capture word patterns in text, such as bigrams (2 words) and trigrams (3 words).

Question 9

What is the Bag of Words (BoW) model?

Answer 9

BoW is a representation that treats text as a collection of words, ignoring grammar and word order, focusing only on word frequency.

Question 10

Describe TF-IDF and its purpose.

Answer 10

TF-IDF (Term Frequency-Inverse Document Frequency) highlights important words in a document by balancing term frequency with rarity across all documents.

Question 11

What is Text Classification?

Answer 11

Text classification is the task of categorizing text into predefined labels, like spam detection or sentiment analysis.

Question 12

Why are Decision Trees used in NLP?

Answer 12

Decision trees are used for text classification tasks due to their simplicity and interpretability, breaking down decisions based on word features.

Question 13

What is the Naïve Bayes algorithm’s role in NLP?

Answer 13

Naïve Bayes is a probabilistic algorithm used for text classification, assuming feature independence to simplify calculations.

Question 14

Explain Zero Counts and Smoothing in Naïve Bayes.

Answer 14

Smoothing assigns a small probability to unseen words in the training data, preventing zero probabilities in calculations.

Question 15

What is Model Evaluation in text classification?

Answer 15

Model evaluation measures a model’s accuracy and reliability on tasks, often using metrics like precision, recall, and F1-score.

Question 16

Define Accuracy in classification.

Answer 16

Accuracy is the percentage of correctly classified instances in a test set, showing overall model correctness.

Question 17

What is Precision in classification?

Answer 17

Precision is the fraction of true positive predictions out of all positive predictions, indicating model reliability in predicting a specific class.

Question 18

What is Recall in classification?

Answer 18

Recall is the fraction of true positives retrieved from all actual positives, measuring model sensitivity to finding relevant items.

Question 19

Why is the F1-Score used?

Answer 19

The F1-score combines precision and recall into a single metric, useful when dealing with imbalanced datasets.

Question 20

What is a Confusion Matrix in model evaluation?

Answer 20

A confusion matrix is a table showing true vs. predicted classifications, helping to visualize model performance across classes.