Sentiment analysis pipeline in NLP - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is the main goal of a sentiment analysis pipeline?

The main goal is to automatically identify and classify the emotional tone (positive, negative, or neutral) expressed in text data.

beginner

Name the typical steps in a sentiment analysis pipeline.

Typical steps include: 1) Text collection, 2) Text cleaning and preprocessing, 3) Feature extraction, 4) Model training or prediction, 5) Evaluation of results.

intermediate

Why is text preprocessing important in sentiment analysis?

Preprocessing cleans the text by removing noise like punctuation, stopwords, and normalizing words, which helps the model understand the true sentiment better.

intermediate

What is a common method for feature extraction in sentiment analysis?

A common method is converting text into numerical vectors using techniques like Bag of Words, TF-IDF, or word embeddings (e.g., Word2Vec, GloVe).

intermediate

How do we evaluate the performance of a sentiment analysis model?

We use metrics like accuracy, precision, recall, and F1-score to measure how well the model predicts sentiment classes compared to true labels.

Which step comes first in a sentiment analysis pipeline?

A Text collection

B Model training

C Feature extraction

D Evaluation

Text collection is the first step to gather the data before any processing or modeling.

What does TF-IDF stand for in feature extraction?

A Term Frequency - Inverse Document Frequency

B Total Frequency - Indexed Document Frequency

C Text Frequency - Inverse Data Frequency

D Term Factor - Indexed Document Factor

TF-IDF measures how important a word is in a document relative to a collection of documents.

Which metric balances precision and recall in model evaluation?

A Accuracy

B Confusion matrix

C F1-score

D Loss

F1-score is the harmonic mean of precision and recall, balancing both.

Why remove stopwords during preprocessing?

A They are misspelled words

B They are rare words

C They add important sentiment meaning

D They are common words that do not add meaning

Stopwords like 'the', 'is', 'and' are common and usually do not help in understanding sentiment.

Which model type is commonly used for sentiment classification?

A Linear regression

B Neural networks

C Decision trees

D K-means clustering

Neural networks, especially deep learning models, are widely used for sentiment classification.

Describe the full process of a sentiment analysis pipeline from raw text to sentiment prediction.

Think about how you would turn a sentence into a sentiment label step-by-step.

Explain why feature extraction is necessary in sentiment analysis and name two common techniques.

Models need numbers, not words, to learn patterns.