Text Preprocessing DATA-SCIENCE-OPTIMIZE Practice Questions (Page 4)

Questions 16–20

1. 16foundation · easy

   What is a common approach to handling frequent words in text preprocessing?

   • AKeeping them as they often carry important meaning
   • BAlways removing them
   • CReplacing them with a special token
   • DConverting them to numbers
   Reveal Solution Discussion Select an answer first
2. 17foundation · easy

   What is the purpose of expanding contractions like 'don't' to 'do not' in text preprocessing?

   • ATo split text into individual tokens
   • BTo remove stop words from the text
   • CTo convert text to lowercase
   • DTo standardize the text and reduce ambiguity
   Reveal Solution Discussion Select an answer first
4. 18foundation · easy

   What is a potential downside of lowercasing all text?

   • AIt makes stemming unnecessary
   • BIt can lose information such as proper nouns or emphasis
   • CIt increases the vocabulary size
   • DIt removes stop words automatically
   Reveal Solution Discussion Select an answer first
5. 19expert · hard

   A team is preprocessing tweets for a hate-speech detection model. Tweets contain hashtags (#racism), mentions (@user), URLs, and emojis. The team wants to reduce noise while preserving sentiment cues. What is the best strategy?

   • AReplace URLs and mentions with placeholders, keep hashtags as tokens, and keep emojis as tokens.
   • BConvert all emojis to their text descriptions (e.g., '😡' to 'angry face').
   • CRemove all URLs, mentions, hashtags, and emojis.
   • DRemove only URLs and keep everything else as-is.
   Reveal Solution Discussion Select an answer first
6. 20foundation · easy

   What does text normalization typically involve?

   • ARemoving all punctuation from the text
   • BConverting text to a bag-of-words representation
   • CApplying consistent formatting such as standardizing whitespace and encoding
   • DSplitting text into individual tokens
   Reveal Solution Discussion Select an answer first