Document-term matrix in NLP - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is a Document-term matrix (DTM)?

A Document-term matrix is a table that shows how often each word appears in each document. Rows are documents, columns are words, and the cells have counts of word appearances.

beginner

Why do we use a Document-term matrix in text analysis?

We use a Document-term matrix to turn text into numbers so computers can understand and analyze it, like finding patterns or training machine learning models.

beginner

What does each row and column represent in a Document-term matrix?

Each row represents a document, and each column represents a unique word (term) from all documents.

intermediate

How can the values in a Document-term matrix be weighted besides simple counts?

Values can be weighted using methods like TF-IDF, which gives more importance to words that are common in one document but rare across others.

intermediate

What is a common problem with Document-term matrices and how is it handled?

DTMs can be very large and sparse (mostly zeros). We handle this by removing rare words, using dimensionality reduction, or applying sparse matrix storage.

What does a cell value in a Document-term matrix usually represent?

A The length of the document

B The number of times a word appears in a document

C The number of documents containing the word

D The total number of words in all documents

Each cell shows how many times a specific word appears in a specific document.

In a Document-term matrix, what do the rows represent?

A Words

B Paragraphs

C Sentences

D Documents

Rows correspond to individual documents in the collection.

Which technique can improve the usefulness of a Document-term matrix by weighting words?

A TF-IDF

B Clustering

C Normalization

D Tokenization

TF-IDF weights words based on their importance across documents.

What is a common issue with Document-term matrices?

A They are sparse with many zeros

B They contain too many images

C They are always too small

D They cannot be used for machine learning

DTMs often have many zeros because most words do not appear in most documents.

How can we reduce the size of a Document-term matrix?

A By adding more documents

B By increasing the number of words

C By removing rare words and using dimensionality reduction

D By converting text to uppercase

Removing rare words and applying techniques like PCA help reduce matrix size.

Explain what a Document-term matrix is and why it is useful in text analysis.

Think about how text is turned into numbers for computers.

Describe common challenges with Document-term matrices and how to address them.

Consider what happens when many words appear rarely.